ArticlePDF Available

A step further – optimizing the natural apple tree habit with the Salsa tree training concept

Acta Hortic. 1130. ISHS 2016. DOI 10.17660/ActaHortic.2016.1130.22
XXIX IHC – Proc. Int. Symposia on the Physiology of Perennial Fruit Crops and
Production Systems and Mechanisation, Precision Horticulture and Robotics
Eds.: D.S. Tustin et al.
A step further – optimizing the natural apple tree habit
with the Salsa tree training concept
Extension service, Route de Mollégès, 13210 St Rémy de Provence, France; 3Aquifruit, Techniques et
Développement Extension Service, 2200 Ch Principal, « Le lion d’Or » Route de Beaupuy, 47200 Marmande,
There is no straightforward relationship between tree density and economic
performance. If the general trend towards higher planting densities has led to
successful results, it has also been ascertained that knowledge on the biology of the
tree and on its reactions to pruning and training procedures is still crucial. The search
for the best compromise between tree manipulations to maintain a given canopy
shape and the minimizing of adverse vegetative reactions has now provided rather
satisfactory answers. As a whole, utilizing a specific tree shape has the advantage of
giving simple objectives and clear rules to implement in orchards. However this may
delay entrance into production and does not improve regularity of bearing. This is
typically what happens when the objective is to establish conic- or cylindric-shaped
tree with a single trunk. Our experience on free-standing apple trees over the past 20
years in France has shown that there is a clear interest in using the strong branches
that may naturally develop from the bottom of the tree. These scaffold branches
behave as intermediate structures between the trunk and fruiting branches and
maintain a better distribution of vegetative growth within the canopy along with
better light distribution to the fruiting sites. Based on these observations we have
proposed some simple rules to train the tree. This canopy system, known as Salsa, is
increasingly interesting to growers due to its satisfactory yield quantity and quality
with lower labour inputs, due to less time required for training and pruning to shape
the tree.
includespruningprocedures,i.e.,theart andscienceofcutting awaya portionofplantfor
horticulturalpurposes,and training,i.e.,the managementoftheoveralltreeshape(Ferree
andSchupp,2003;Robinson,2003).Thesetwoscalesareintimately related, and for
hasadirectimpactontreearchitectureand size.Onthe otherhand,therootstock‐cultivar
the world shows that the grower’s choice is a compromise among several constraints.
Indeed, the natural environment, soil and climate, and the prevalence of some pests and
pattern. The socio‐economic context also has a great influence duetotherelativecostof
orchard management, especially for training, pruning and harvesting in a given region,
Inthispaper,wepresenttheworksthathavebeendeveloped overthepast20years
pruning.Inasecondpartwewillreportontheevolutionoftraining over thepast10years
whichaimsto combinefruitqualityand regularyieldwitha fastentranceintoproduction
andareductionofthe managementcosts.TheSalsatrainingsystemisbasedoni)abetter
useofthenaturaltreearchitectural framework,especiallyduringthe veryfirstyearsafter
plantation,andii)abetterdistributionofthefruitingshoot,i.e.,thebourseand itsbourse‐
developing training and pruning methods that constrain the treehasbeenadvocatedby
manyhorticulturistsoverthepastdecades(Lespinasse,1977,1980; Forsheyetal.,1992).
with fruiting in terminal position on long shoots, i.e. Type IVaccordingtoLespinasses
typology, have a more regular fruiting pattern (Lespinasse, 1992; Lespinasse and Delort,
1986;Laurietal., 1995). Thisgeneralrelationshipbetweenthe overalltreeshape andthe
fruitingpatternhasbeenrelatedatthebranchscaletotherelationship between shoot
lengthandreturn‐bloom(LauriandLespinasse, 1993; Lauri andTrottier, 2004).However,
Lespinassestypologyandischaracterizedbyanalternatebearing pattern (Lauri et al.,
1997b). Therefore, all relations between the whole‐tree typology and the fruiting pattern
‐ the establishment of a multi‐trunk tree when the cultivar intrinsically has this
‐ thedisplayoffruitingshootswithinthetreecanopywithouttheneedforfruiting
From the single- to the multi-trunk tree
Maintainingauniquetrunkisoftenartificialbecausethetree intrinsically develops
sylleptic shoots concurrent to the trunk in the year of plantation (Figure 1a). This is
grafted on dwarfing rootstocks such as M.9. It also varies depending on the cultivar. This
phenomenon is usuallynotobservedbythegrowerplantingone‐ortwo‐year‐old grafted
trunk and to encourage proleptic branching higher on the trunk inthefollowingyear.
Figure1. Unpruned3‐year‐old‘GrannySmith’treegraftedonM7(a),andsilhouetteofa5‐
From the fruiting branch to the fruiting shoot
trunkandbearingthefruitingshoots,hasbeenproposedasaway to balance vegetative
The centrifugal training system was a further step, with the integration of the light well
artificial extinction (also called extinction pruning) to improvethefunctioningofthe
yearsto gainadeeper knowledgeofthe vegetativeand floweringandfruitingstrategiesat
thefruitingbranchscale(LauriandLespinasse,1998,2001;Lauri et al., 2004; Lauri and
Laurens,2005;Lauri,2008).Amainresultwasthatregularbearing at the level of the
fruiting shoot (bourse‐over‐bourse) consistently improved if branching density decreased
Delort,1993).Thisconceptofautonomyofthe fruitingshootwithregard toreturnbloom
Lespinasse,1993;Laurietal.,2004).Aconsequenceofthesefindingsis that although the
ThecurrenttrainingparadigmimplementedinSalsatrained treesistopromotethe
has a lesser importance. The objective is not to build a tree with a rigid and often time‐
the best positioning of the fruiting shoots in the tree canopy andtotrainthestructural
theplantmaterialandtheenvironmentisnotrelevant.TheSalsaconceptis basedontwo
therootstock/cultivarcombinationthatisgrownareofhighimportance to adapt these
‐ At the whole‐tree scale, all branches at the bottom of the tree, usually 2 to 5,
dependingontherootstockcultivar combination and the environment, are kept.
Dependingonthegrower,theplantmaterialfromthenurseryor the region (e.g.,
branchesfrom50cmupwardarekept.Thesebranchesareattached with cordsto
horizontalwiresinaflexiblewayalongtherowoneachsideof the row. The
objective is to develop upright or slanted reiterative trunks on which the fruiting
structuralframeworktofillthevolume intherowasfastas possible.Althoughthe
round‐shaped tree has repeatedly been shown to be the best crown geometry for
light interception (Smith et al., 2014) but also for the balance between vegetative
about3mhighand0.81mwidth,withplantingdistancesof3.5×1.2‐1.5 m
corresponding to ca. 2000‐2400 trees ha‐1orca.6000reiterativetrunksha
‐1. In
similarheightasabovebutwithalowerplantingdensity,i.e., 4.2‐4.5×1.5‐1.8 m
TheSalsa‐trainedtreemay beobtaineddirectlyinthegrower’sorchardbyplanting
ispatentedbysome nurseries (e.g., VivaiMazzoniNurseries,Italy) andsoldunder
Salsasystemdoesnotbelongtothepalmettesystem,thislatter one being mainly
ThefruitingbranchconceptisabandonedintheSalsasystem. As a consequence,
branchbending,whichisatime‐consumingoperation,is consistently reducedand
shoots are directly attached along the reiterative trunks and are distributed
homogeneouslywithinthetreecanopyvolume.Toimprovetheautonomy of the
fruiting shoots the porosity of the canopy is obtained through both artificial
thinningcutscarriedouton2yearoldandolderbranchesinovercrowded and
shadedsitesinthecanopy. Oncethefruiting shootispositioned withinthecanopy,
Our experience with the Salsa system in growers orchards over the last 10 years
generated some valid points. From the economic point of view, theSalsasystemreduces
investment at planting and during the orchard life‐span leadingtoafasterreturnon
oabetterbalancebetweenvegetativegrowthandfruiting,especially in conditions
oanimprovedaccesstothetreeandaneasieruseofplatformsfor treetrainingand
The apple tree ideotype is discussed in several papers especiallyinrelationtotree
al., 2014). However, a tree ideotype is only one element of thepuzzlethatleadstothe
economic success of the orchard. The symbiosis between the genetically‐determined tree
ideotypeandtheculturalideotype,thislatteroneincludingthe pruning and training
with mild winter and/or hot and sunny summer brings valuable information on how to
improve the pruning and training to maintain high and regularly fruiting in adverse
Literature cited
Ferree,D.C., andSchupp,J.R. (2003).Pruningandtraining physiology. InApples:Botany,ProductionandUses,
IPCC.(2013).Summary forpolicymakers.In ClimateChange2013: The PhysicalScience Basis. Contributionof
Lauri,P.E. (2008).Trends in appletrainingin France–anarchitecturalandecophysiological perspective.Acta
atINRA,France‐contributiontotheideotypeapproach.ActaHortic. 663, 357–362
Lauri, P.E., and Laurens, F. (2005). Architectural types in a pple (MalusXdomestica Borkh.). In Crops: Growth,
Lauri, P.E., and Lespinasse, J.M. (1993). The relationship between cultivar fruiting type and fruiting branch
Lauri,P.E.,and Lespinasse,J.M.(2001). Genotypeofapple treesaffectsgrowthandfruitingresponsestoshoot
Lauri, P.E., Térouanne, E., Lespinasse, J.M., Regnard, J.L., and Kelner, J.J. (1995). Genotypic differences in the
Lauri,P.E., Lespinasse,J.M.,and Térouanne,E. (1997a). Vegetative growthandreproductivestrategiesin apple
fruiting branches ‐ An investigation into various cultivars. Acta Hortic. 451, 717–724
Lauri, P.E., Térouanne, E., and Lespinasse, J.M. (1997b). Relationship between the earlydevelopmentofapple
Lauri, P.E., Willaume, M., Larrive, G., and Lespinasse, J.M. (2004). The concept of centrifugal training in apple
aimed at optimizing the relationship between growth and fruiting. Acta Hortic. 636, 35–42
Lauri, P.E., Hucbourg, B., Ramonguilhem, M., and Méry, D. (2011). An architectural‐based tree training and
pruning–identificationofkeyfeaturesintheapple.ActaHortic. 903, 589–596
Lespinasse, J.M. (1980). La Conduite du Pommier. II. L’Axe Vertical, la RénovationdesVergers(2ème partie)
Lespinasse, J.M. (1996). Apple orchard management practices in France.From the vertical axis to the solaxe.
Lespinasse, J.M., and Delort, F. (1986). Apple tree management in vertical axis: appraisal after ten years of
Looney, N.E., and Lane, W.D. (1984). Spur‐type growth mutants ofMcIntoshapple:areviewoftheirgenetics,
Smith, D.D., Sperry, J.S., Enquist, B.J., Savage, V.M., McCulloh, K.A., and Bentley, L.P. (2014). Deviation from
symmetrically self‐similar branching in trees predicts altered hydraulics, mechanics, light interception and
a new concept for crop load regulation. Acta Hortic. 932, 195–202
... A first lesson is, therefore, to be able to define a good pruning strategy, not only to control tree size at the desired height and volume but also to avoid excessive pruning that could entail vigorous regrowth, which would be conducive to pest infestation and may put fruiting at a disadvantage. Various training systems have been proposed in the past to combine high fruiting quality and efficient pest management (Lauri, 2008;Lauri et al., 2009;Tustin et al., 2011) through good control of the homogeneity of shoot vigour (Lauri et al., 2016a). ...
Full-text available
In this chapter, we first describe and illustrate various agronomic IPM practices (Section 2) and plant diversity-based practices (Section 3) with detailed examples from fruit production, mainly on pome and stone fruit. We then provide some insights on the design of cropping systems that combine the above-mentioned IPM practices at field, farm and agri-food system scales (Section 4). We also show that designing such systems calls for additional research and new approaches.
... A first lesson is, therefore, to be able to define a good pruning strategy, not only to control tree size at the desired height and volume but also to avoid excessive pruning that could entail vigorous regrowth, which would be conducive to pest infestation and may put fruiting at a disadvantage. Various training systems have been proposed in the past to combine high fruiting quality and efficient pest management (Lauri, 2008;Lauri et al., 2009;Tustin et al., 2011) through good control of the homogeneity of shoot vigour (Lauri et al., 2016a). ...
IPM promotes the design of resilient systems that reduce both pest attacks and damage to cultivated plants. It is based on practices that foster plant tolerance and the control of pests by their natural enemies. The challenge is to combine those practices in a coherent system that should associate the benefits or even develop synergies among their identified partial effects and that should be operable for the farmer. The IPM practices considered in this chapter are related to a) the crop and its annual management; b) the enhancement of plant diversity in the cropping system at the field scale using companion plants and intercropping, as well as diversification with other cash crops. The chapter also provides c) some insights on the design of cropping systems that combine the above-mentioned IPM practices at field, farm and agrifood system scales. It also shows that designing such systems calls for additional research and new approaches.
... Second, it improves light penetration within the tree canopy Breen et al., 2016b), optimizing fruit-set (Breen et al., 2014(Breen et al., , 2016a, return-bloom and fruit quality . Although artificial extinction has to be applied on well-balanced trees, that is without strong heading-back on vigorous shoots, it is successfully applied on various tree shapes, whether with a single trunk such as the tall spindle or centrifugal training (Tustin et al., 2009;Lauri et al., , 2007 systems, or with several reiterative trunks such as the Salsa system (Lauri et al., 2016a). ...
Developing sustainable apple cultivation is based on both a better knowledge of tree architecture and physiology in relation to fruiting, and on how the tree interacts with its abiotic and biotic environments. Improving knowledge in these domains is crucial to take into account the societal demand towards less input-dependent orchards. This chapter provides an overview of apple tree growth and fruiting, exploring the bases for sustainable apple training and pruning management. The chapter also challenges the current apple production agroecosystem and looks ahead to future research trends in this area.
... This new vision also lends further support to the concept that, from the practical point of view, efficient manipulations of source and sink organs are far more important than giving the tree a standardized shape (Lauri and Corelli-Grappadelli, 2014). Indeed, artificial extinction is successfully applied on various tree shapes, central leader tall spindle system , centrifugal training (Tustin et al., 2009;Lauri et al., , 2007 or Salsa (Lauri et al., 2016a). ...
Conference Paper
The majority of the research in apple tree architecture and physiology over the past decades is based on trees grown in mono-cultivar high-density orchards. Here we propose to connect knowledge on the tree itself with knowledge on the system in which it is grown. In both cases, scientific advances are intrinsically connected with practical aspects. It is first shown that although whole-tree shape variability presents a descriptive interest it is not strictly related to the fruiting pattern, which is a major issue for apple cultivation. Architectural analyses have been developed not only to gain knowledge on the dynamics of tree growth and branching but also to analyse thoroughly the physiological conditions for an optimized fruit-set, fruit quality, and return-bloom. Shoot length and the physiological abortion of young shoots on the same branch (extinction) have shown to be two main components for regular yield and high-quality fruiting. These architectural analyses have also been at the origin of an innovative pruning procedure, artificial spur extinction, developed under the paradigm of precision horticulture. It is implemented in various training systems. The sustainability of intensive apple growing systems is now questioned, especially because of their high dependency on external inputs whether phytosanitary products or fertilizers. Agroforestry, i.e., growing woody perennials and annuals in spatial mixtures and/or temporal sequences, is proposed as a means to reduce such dependencies. It uses plant biodiversity to enhance ecological services including multiple cropping. For apple, the social-economic acceptability and performance of such systems have to be investigated. From the horticultural point of view, this new context challenges the choice of plant material and of training and pruning concepts developed in intensive orchards. From the scientific point of view, apple tree-based agroforestry systems address architectural and physiological issues on how the apple tree can develop and fruit satisfactorily in such biotic and abiotic contexts.
... This new vision also lends further support to the concept that, from the practical point of view, efficient manipulations of source and sink organs are far more important than giving the tree a standardized shape (Lauri and Corelli-Grappadelli, 2014). Indeed, artificial extinction is successfully applied on various tree shapes, central leader tall spindle system , centrifugal training (Tustin et al., 2009;Lauri et al., , 2007 or Salsa (Lauri et al., 2016a). ...
... Second, it improves light penetration within the tree canopy Breen et al., 2016b), optimizing fruit-set (Breen et al., 2014(Breen et al., , 2016a, return-bloom and fruit quality . Although artificial extinction has to be applied on well-balanced trees, that is without strong heading-back on vigorous shoots, it is successfully applied on various tree shapes, whether with a single trunk such as the tall spindle or centrifugal training (Tustin et al., 2009;Lauri et al., , 2007 systems, or with several reiterative trunks such as the Salsa system (Lauri et al., 2016a). ...
Conference Paper
Full-text available
The procedures that are given to maintain the apple tree architecture within a given “shape” do not suffice in themselves to fulfill the two main objectives of apple tree cultivation that are regular bearing and fruit quality. The intrinsic architectural and functioning characteristics of the cultivar-(interstock)-rootstock entity are determinant on the success of the orchard. In the following, I will take the exampIe of applied research works developed in France to illustrate that the progress in apple orchard performance over the last 40 years was essentially based on the improvement of our knowledge in apple tree architecture and functioning.
Full-text available
Growth of individual apple fruit during their early development occurs in an environment of intense competition among individuals within a spur as well as among adjacent sinks. To increase fruit size and quality, chemical and hand thinning are used to reduce fruit numbers on trees. Thinning is essential but expensive and chemical thinning responses are often unpredictably variable. Thinning can also be considered to be wasteful of dry matter resources because many young fruit are removed from the tree. This paper examines the concept of regulating floral bud distributions within the tree, in order to manipulate fruit set and early fruit development to more optimally use dry matter resources at this time of limited supply and intense competetion. In a randomized block layout, artificial bud extinction was used, just prior to bud break, to set floral bud densities in a range from 2 to 6 buds per cm2 branch cross-sectional area on tall spindle trees of the heavy-flowering apple cultivar 'Scifresh' (6 years old on M.9 rootstock). Equivalent unmodified trees (controls) were thinned by hand after final drop to fruit numbers equal to those calculated for bud extinction treatments. No chemical thinning was used. Fruit set and return bloom were measured annually during three successive years of treatment. The patterns of fruit set showed that the set of individual buds was highly dependent on the total density of floral buds present and that responses were repeatable annually. In conditions of very high floral bud density, fruit set of individual buds was low and almost 60% of buds failed to set any fruit. Reducing floral bud density by artificial bud extinction increased the proportion of floral buds that set fruit and increased the number of fruit set on individual buds. Annual bud extinction treatments induced a high proportion of buds into return bloom each spring. These fruit set and return bloom responses show artificial bud extinction could be a valuable physiological tool for optimizing fruit development and crop load. Bud extinction prior to growth resumption in spring largely sets the target fruit number and controls the number of competing floral sites during the early season when reserves and current resources are most limited. In this way the process appears to facilitate allocation of resources at times of limited supply to sites whose fruit are already selected for development to maturity.
This book, with contributions from 40 scientists from 8 countries, summarizes the current research information on apples and their culture, and will be of value to horticultural students, research and extension workers, and professional fruit growers. The 24 chapters are presented in 6 parts. Part I, the introductory section, considers taxonomy, and world production and trade. Parts II-VI cover the following aspects: Plant materials (breeding and genetics, cultivars, rootstocks, and propagation and planting stock); Apple physiology and environmental influences (flowering and fruiting, water relations, light relations and temperature); Orchard and tree management (site selection and orchard establishment, nutritional requirements, orchard floor management, pruning and training, planting systems, thinning and the vegetative:fruiting balance, and endogenous growth regulators and growth regulator application); Crop protection (diseases, ecology and pest management, frost protection, integrated fruit production and organic production); and Harvesting, handling and utilization (fruit maturity and ripening, harvesting and handling, storage, physiological and pathological disorders, and production and handling techniques for processing apples).
Les évolutions depuis 30 ans conduisent à proposer des concepts de gestion de l'arbre en verger plus respectueux de sa physiologie
The new apple cultivar 'Scifresh' has several genotypic traits that impair commercial productivity and fruit quality. This study is on the consequences of the e×ceptional floral precocity of 'Scifresh' trees, where almost all terminal, spur and a×illary buds flower annually, producing many weakly-developed floral spurs. Observed low fruit set, high sensitivity to chemical thinning and high a×illary bud and spur e×tinction are thought to be responses to 'Scifresh' floral behaviour and may be related to competition for resources during early fruit development. Changes in tree management to improve resource allocation to floral spurs early in seasonal development may enhance fruit set and fruit development, thereby increasing productivity, fruit size and quality. Centrifugal Training (CT) tree management, which regulates the density of fruiting sites on branches, was investigated to alter the early-season physiological status of 'Scifresh'/M.9 trees to improve both cropping and fruit quality. At budbreak, spurs on branches of CT trees were thinned to numbers calculated to produce 4, 5, or 6 fruit per cm 2 of branch cross-sectional area when cropped using either one (CT1) or two (CT2) fruit per spur. These treatments were compared with standard New Zealand Vertical A×e tree management (VA) thinned to the same crop treatments after final fruitlet drop. Between 50 and 65% of floral spurs on standard VA trees failed to set any fruit and this proportion was reduced to 25-35% in CT1 and to 18-25% in CT2 treatments. The proportion of spurs that set two or more fruit more than doubled in response to CT. Although CT increased fruit set, both crop density and yields at harvest were lower than with standard VA training for equivalent crop treatments. Mean fruit weight declined in response to increasing crop density, although it was further reduced in those treatments cropped as two fruit per spur. Centrifugal Training altered the composition of vegetative annual shoots at the branch unit level, resulting in a ratio of e×tension shoots to spurs of 1.36 compared with 0.39 for VA-trained trees. Node number, internode length and basal diameter of e×tension shoots all increased in response to Centrifugal Training.