2/24/2014
Using Concepts of Shoot Growth and
Architecture to Understand and Predict
Responses of Fruit Trees to Pruning
Ted DeJong
Why prune?
 For training to produce a vigorous, mechanically
strong, framework
 To obtain a well-shaped tree for convenience of
orchard management
 For optimal capture and distribution of sunlight
throughout the canopy
 Promote fruit size (crop reduction technique)
 Stimulate new fruiting wood (bearing wood renewal)
 Regulate crop production over years (decrease
alternate bearing)
Basic concepts for understanding tree
growth and responses to pruning
 Three types of shoots
 Two types of shoot growth
 The basic structure of shoots determines cropping and
subsequent growth
 Duration of shoot growth
 Apical dominance
 Gravitropism
 Carbohydrate/resource balance
 Concept of “reiteration”
 Types of pruning cuts
1
 2/24/2014

Three types of shoots

 Proleptic: shoots that grow out from terminal
or lateral meristems after a dormant bud has
been formed and period of dormancy occurs.
 Syleptic: shoots that grow out from lateral
meristems in the axils of leaves without an
intervening period of dormancy.
 Epicormic: shoots that grow from dormant
preventitious buds on older branches (usually
after a pruning cut, limb break or extreme limb
bend).
Syleptic Proleptic Epicormic

Two types of shoot growth

 Neoformed shoot
growth: shoot growth that
 Preformed shoot growth: grows as a result of the
shoot (or spur) growth production of new nodes
that grows as an that are produced during
extension of internodes the current growing
between nodes that were season.
preformed in proleptic
For shoots starting from
buds (usually limited to proleptic buds this results
6-10 nodes). in “mixed” shoots.

2
 2/24/2014

The basic structure of each shoot type is

relatively consistent and can be
systematically studied and characterized. X1 0 0 0 0 2 2 2 3 3 3 3 2 2 1 1 1 1 0 0

X2 0 0 0 0 0 0 1 2 2 2 2 1 1 0 0 0 0 0 0

X1: type of axillary development

3 sylleptic shoot
2 vegetative bud
1 central flower
0 blind node
X2 : number of lateral flowers
(Fournier et al. 1998)

Watersprouts: 6 states + terminal bud

Initial Probabilities 0.98 0.02

38 –90
Occupancy 5.28 2.43 3.03 2.42 2.15 2.98 nodes
distribution mean
0.63 0.99 0.05 0.49 0.43 0.25
Transition Probabilities 0.37 0.11 0.51 0.04
0.83 0.53 0.25

Observation Distribution 0.50

V>B>S V S V S >B B>F T

Variable 1 0.52>0.28>0.20 0.99 0.97 0.99 0.88>.0.10 0.82>0.13

0 2>1>0 0 2>0>1 0 0
Variable 2
0.96 0.57>0.22>0.21 0.99 0.42>0.36>0.22 0.99 0.99

3
 2/24/2014

Long shoot: 5 states + terminal bud

Medium shoot: 4 states + Terminal bud

Initial Probabilities Initial Probabilities 0.92 0.08

0.92 0.08
Occupancy Occupancy distribution 3.39 9.28 4.64 1.86 12-26
3.50 12.30 5.74 3.89 2.05 21-33 nodes
distribution mean mean
nodes

Transition Probabilities 0.99 0.99 0.70 1.00

Transition probabilities 0.97 0.96 0.91 0.56 1.00
0.30
0.44
Observation Distribution 0.08
Observation Distribution
V F>B=V B T
Variable 1 B>V V V F>B>V B T
Variable 1 B>V>F
0.94 0.78>0.11=0.11 1.00
0.69>0.20>0.11
0.54>0.41 0.99 0.94 0.47>0.46>0.07 1.00
0>1>2 0 0
2>1>0 0>1>2 0 0 0
Variable 2 0>1 Variable 2 0.99
0.60>0.26>0.14 0.99 1.00
0.92>0.05 0.42>0.33>0.25 0.84>0.11>0.05 0.99 1.00

Differences in shoot growth duration are related to

Small shoot: 4 states + terminal bud type of growth, type of shoot and plastochron.
0.89 0.11

8-18 nodes
The average plastochron for peach is about 2-3
4.17 3.04 4.89 2.59
days.
0.60 0.96 0.26 1.00
 Spurs have fewer than 11 nodes, are entirely
0.40 0.74
B>F>V V>F>B F>B B T
preformed and stop growing within 30 days
0.83>0.09>0.08 0.77>0.15>0.08 0.75>0.22 1.00 after bud break. Most short shoots are similar.
 Medium and long shoots are mixed shoots
Spur: 3 states + terminal bud with preformed and neoformed growth and up
0.99
4.45 2.24 5-11 nodes to 34 nodes. They stop growing within 90 days
0.99 1.00
of bud break (by June 15 in California).
 Epicormic shoots are entirely neoformed, have
B>F>V F>B T
0.71>0.20>0.09 0.76>0.23 up to 90 nodes and thus grow until fall.

4
 2/24/2014

Apical dominance: the influence or control of shoot

development and growth of lower buds or shoots by
more apical buds or shoots

Three manifestations of apical dominance

 Correlative inhibition: suppression of
lateral shoot growth by a vigorously
growing apical meristem during the
current season’s growth.
 Apical control: tendency for terminal and
distal lateral shoots to depress the growth
of more basal (subordinate) shoots.
 Shoot epinasty: tendency for actively
growing upper, distal shoots to influence
the branch angle of basal shoots (usually
making them wider).
If shoot is not pruned all new shoots will
be shorter than parent shoot.

Gravitropism

 Gravitropism is the
behavior of growth in
response to gravity.
 Shoot growth displays
negative geotropism.
 Fruit species vary in the
amount of gravitropism
they exhibit (cherries a lot,
peaches less).
 Gravitropic responses can
vary with genotype within a
species.

5
 2/24/2014

Carbohydrate/resource balance Reiteration

“Reiteration” is the tendency for a tree to “replace” any
vegetative part of the tree that is lost or removed
 If allowed to grow naturally, when a tree goes through limb breakage or pruning. In the case of loss
into dormancy its natural shoot growth potential of major pieces of individual units like shoots, the
for the following spring is a function of a balance new shoot will have characteristics similar to the
shoot that is being replaced.
between the carbohydrate and nutrients stored
prior to dormancy and the number of buds.
The “strength” of reiteration after pruning depends on
 Any pruning during dormancy will change this the type of pruning cut, the amount growth that was
balance in favor of stimulating more growth, lost, and the timing of the pruning. The growth of
epicormic shoots from preventitious buds is the tree’s
often in the form of epicormic shoots natural way to rapidly replace parts of the canopy that
(watersprouts) from preventitious buds. is lost.

In orchards where heavy topping is used in California, reiterative

Without pruning the proportion of shoots in the small
growth can be 3 m in one season.
categories increases and fewer long shoots are produced.
Watersprouts are not produced if trees are not pruned unless
there is limb breakage or severe bending.

6
 2/24/2014

Types of pruning cuts

 Heading cuts: the terminal part of the main

axis of the shoot is removed and the basal Application of the concepts
portion remains attached to the tree.
 Thinning cuts: the entire axis of a shoot is
removed from the tree.

In young, vigorous trees the

new growth on a previous
year’s shoot that expressed
strong correlative inhibition
in the first year, we can
expect expression of strong
correlative inhibition in the
second year.
Apical control and shoot
epinasty effects will also be
apparent with new shoots
being progressively shorter
and branching angles
progressively wider from the
apex to the base of the
shoot. This behavior may be partially
because of lower chilling
requirements and earlier
1st year 2nd year growth and flow of auxin from
buds nearer to the apex to the
more basal buds.

7
 2/24/2014

If we go back to the original one-year-old shoot and head it back to
one-half it’s length we have substantially changed things:
•The inherent apical dominance affects have been reduced
The result is that the more distal buds that remain on the shoot will
probably grow longer shoots than the shoots that would have been
produced if the original shoot had been left un-pruned. Note also that
the apical control and shoot epinasty are re-established in the new
branch.

•The new growth will be invigorated because more than half of the
buds have been removed (nodes are closer together in the upper
The reiteration rule tells us
half of the shoot) and the mass and therefore the stored that the new longer shoots
carbohydrates are greater in the basal half of the shoot. will have general
characteristics similar to the
original un-pruned parent
shoot but the more distal
shoots may be longer than the
original parent shoot.

This is the basis for the

general pruning rule: “cut
what you want to grow.”

In peach production heavy pruning leads to major tree management

problems because of excessive stimulation of epicormic shoots.

These epicormic shoots are entirely

neoformed because they are
stimulated from latent preventitious
buds. They are a problem because
they grow until conditions are not
favorable for growth.
We must find ways to minimize
pruning cuts that stimulate the
production and growth of these types
of shoots in mature bearing fruit
trees. This is probably where size-
controlling rootstocks and controlled
water deficits after harvest of early
maturing cultivars can play a role.
However our current training systems
are very dependent on the concept of
reiteration and epicormic shoot
production for the rapid development
of tree canopies.

8
 2/24/2014

Putting the concepts together in the

In California we often use a very
L-PEACH Model aggressive style of pruning to train
young trees.
Goals of the L-PEACH modeling project: The tree on the right is a simulation of
a vigorous tree at the end of the
 Build a model of architectural tree growth, carbon season in a nursery.
assimilation and transport, and dry matter partitioning
based on developmental and physiological concepts
of how trees grow
 Model tree and fruit growth from the organ to whole
plant level
This tree would be transplanted into
 Visualization and quantification of plant growth the orchard and pruned to ~0.5 m with
 Model responses of tree growth to the environment all of the side shoots removed. Why
and to horticultural management practices do we remove all the side shoots?

 I will use the model to demonstrate how these concepts of

tree growth are used in our current tree training practices.

Unpruned syleptic shoots from the

previous year only produce shoots
from proleptic buds that are weaker
than the parent shoot. If these are
removed more “water sprouts” are
produced.

Pruned3years L-Peach-d.wmv

9

In the first year in the orchard that
tree develops several strong
branches from water sprouts as a
result of the strong pruning at the
time of planting. These scaffolds are
mainly epicormic shoots
(watersprouts) stimulated by heavy
pruning.
In the following dormant season the
grower selects 2, 3 or 4 strong
branches to develop the primary
scaffolds and again heads them to
stimulate more strong growth (water
sprouts) to either continue the
selected scaffolds or develop multiple
secondary scaffolds.
Later we would like to produce
more “mixed shoots” and fewer
or no water sprouts.
The L-PEACH model can
simulate a reasonable in silico
representation of a peach tree
that can be pruned to any
configuration normally used in
California peach orchards.
2/24/2014
In the second year in the orchard the
tree again responds to the heavy
pruning and develops multiple
branches (water sprouts) on the ends
of the primary scaffolds.
In the following dormant season the
grower again selects branches to
either reinforce the previously
selected scaffolds or develop multiple
secondary scaffolds.
Two other features that are
implemented in L-PEACH that
are a form of pruning are
simulated topping (a practice
that is increasingly being used
by growers to manage pruning
costs) and fruit thinning.
Fruit thinning can be done
manually (as in the orchard) by
selectively removing individual
fruit or automatically by
specifying the date of thinning
and the minimum distance
(number of metamers) between
fruit at the beginning of a
simulation.
10
 2/24/2014

Thank you for your attention

Conclusions
 It is possible to take the “art” out of pruning by
analyzing and integrating the factors that regulate
shoot growth and development.

 Now that we have a better understanding of how

peach trees actually grow we need to apply that
knowledge to find effective means to minimize
excessive tree growth to reduce labor costs and direct
more carbohydrates and nutrients to fruit production.

Collaborators
UC Davis Univ. of Calgary
Yaffa Grossman Przemyslaw Prusinkiewicz
Mitch Allen Radek Karwowski
Romeo Favreau Pavol Federol
Claudia Negron Brendan Lane
Colin Smith Mik Cieślak
David Da Silva INRA, Montpellier
Eric Walton Evelyne Costes
Edelgard Pavel Yann Guedon
Mehdi Ben Mimoun
Boris Basile Wageningen UR
Jordi Marsal Jan Goudriaan
Scott Johnson IRTA, Lleida
Kevin Day Gerardo Lopez
Steve Weinbaum Inigo Auzmendi

11