CFPACanning Fruit Producers' Assoc.

Submit to:

Wiehahn VictorPO Box 426Paarl, 7620

Tel:+27 (0)21 872 1501

inmaak@mweb.co.za

DFPTDeciduous Fruit Producers' Trust

Submit to:

Louise LiebenbergSuite 275, Postnet X5061

Stellenbosch, 7599Tel: +27 (0)21 882 8470/1

louise@dfptresearch.co.za

DFTSDried Fruit Technical Services

Submit to:

Dappie SmitPO Box 426Paarl, 7620

Tel:+27 (0)21 872 1501

dappies@dtd.co.za

Winetech

Submit to:

Jan BooysenPO Box 825Paarl, 7624

Tel: +27 (0)21 807 3324

booysenj@kwv.co.za

X

FINAL REPORTFOR 2004/2005

PROGRAMME & PROJECT LEADER INFORMATION

Title, initials, surnamePresent positionAddressTel. /Cell no.FaxE-mail

Programme leader Project leaderDr J SteenkampManager: ViticultureARC Infruitec-Nietvoorbij021 809 3200021 809 3002steenkampj@arc.agric.za

Dr JJ HunterSpecialist ScientistARC Infruitec-Nietvoorbij021 809 3057021 809 3002hunterk@arc.aqric.za

PROJECT INFORMATION

| Project number ,| WW 12/23

Project title Determination of optimal grape and wine quality of Shiraz/Richter 99 andthe relationship with seasonal variation in leaf and berry metabolism, asaffected by microclimate.

CFPADFPTDFTSWinetech

Other

Viticulture : Cultivation & Optimal grape composition fordifferent wine objectives

Fruit kind(s) Winegrapes

Start date (dd/mm/yyyy) 01/04/2001 End date (dd/mm/yyyy) 31/03/2005

FINAL SUMMARY OF RESEARCH PROJECT

PROGRAMME & PROJECT LEADER INFORMATION

Title, initials, surnameInstitutionTel. / Cell no.E-mail

Programme leaderDr J SteenkampARC Infruitec-Nietvoorbij021 809 3200steenkampj@arc.agric.za

Project leaderDr JJ HunterARC Infruitec-Nietvoorbij021 809 3057hunterk@arc.agric.za

PROJECT INFORMATION

Project number WW12/23

Project title Determination of optimal grape and wine quality of Shiraz/Richter 99and the relationship with seasonal variation in leaf and berrymetabolism, as affected by microclimate.

Fruit kind(s) Winegrapes

Start date(dd/mm/yyyy)

01/04/2001 End date(dd/mm/yyyy)

31/03/2005

Research was done on a Shiraz/R99 vineyard in the Stellenbosch region with the purpose ofdefining environmental, canopy and grape parameters that may be suitable as eventualpractical indicators for obtaining particular styles of grapes and wine. During the course ofthe project, several facets of this aspect of cultivation were investigated, more often incollaboration with Spanish, French, Italian and Swiss researchers. Furthermore, one PhDstudy (University of Palermo, Italy) and one MSc study (University of Stellenbosch, SouthAfrica) were completed as part of the project. The summary of only the main focus of theproject is presented. Summaries of the rest of the facets are presented in the "Results andDiscussion" of the report.

Optimal ripeness is defined according to the style of wine that is required. The latter isultimately dictated by the market. Soil and climate may have a dictating effect on typicalexpression of wine. The level of grape and wine quality achieved and the potential forobtaining different styles of wine are determined by the integrated effect of the naturalcharacteristics of the terroir and technological intervention (long and short term cultivationpractices). The growth conditions that the grapevine is subjected to should allow optimalmetabolic activity in roots, permanent structure, canopy and grapes and the potential forthese organs to develop and support each other until the desired grape quality and style isreached. Monitoring of morphological and physiological parameters in the canopy andgrapes, ultimately displaying the integrated effect of the growth environment, is critical in ourquest for finding indicators that may be associated with a particular grape and wine style.This has not previously been investigated systematically.

Vines were vertically trellised and spaced 2.75 x 1.5 m in north-south orientated rows on aGlenrosa soil and a west-facing slope. Microsprinkler-irrigation was applied at pea berry sizeand at veraison stages. The 1.4 m canopies were shoot-positioned and topped. Fortnightlysampling was done from berry set up to two weeks post-veraison, after which harvesting forwine making was done approximately every four days. Microclimate, vegetative,reproductive and physiological parameters were investigated and changes during alcoholic

fermentation monitored at each harvesting stage. Wines were made and analysed.Similarities in patterns as well as various ratios between the different parameters wereinvestigated. Results are argued against canopy performance, carbon allocation, waterrelations, production level, and sugar, acidity, anthocyanin, phenolic and tannin contents ofthe grapes as well as wine quality and composition. Ratios for potential practical use indetermining optimal grape quality, time of harvesting and expected wine style are discussed.Ideally, the vineyard growth conditions should allow the prediction and distinction of differentharvesting times, representing different styles of wine that can then either be madeunilaterally or multilaterally by means of blending. Irrespective of the style of wine required, itis important that grapes are always harvested in a physicochemical state that wouldguarantee maximum (optimal) quality of the product for which they are intended. Cultivationpractices should be applied in such a way that the full grape quality potential of the vine isexpressed on the particular terroir. If the terroir and cultivation practices do not allow normalphysiological performance and grape potential to be fully expressed, as is often the case inpractice, wine quality and style options will be reduced. Judicious selection of terroir andlong-term practices is therefore critical.

The study clearly showed that optimal ripeness can not simply be described by the maximumaccumulation of grape components, but rather represents a complex, particular combinationof physiological/biochemical changes (in leaves and grapes in particular growth phases),physical changes (homogeneity of vine and canopy structure as well as bunch and berrysize), and required grape/wine style and market preferences.

In this study, optimal harvesting time for a particular style of wine could be illustrated byusing classic parameters of which information can be easily obtained during the growthseason in the vineyard and in the winery by producers and winemakers. Positive indicationsof global applicability of the wine quality and style indicators found in this study (grapeparameters and ratios) were evident.

Final report 4

FINAL REPORT

1. Problem identification and objectives

PROBLEM ADDRESSED/EVENTUAL OBJECTIVE1. To increase grape quality.2. To determine optimal harvesting time for a particular wine style.3. To determine the importance of different seasonal growth phases in the establishment

of grape and wine quality.4. To obtain practical parameters for manipulation and judging of optimal grape and wine

quality.

2. Workplan (materials & methods)

Included in publications.

3. Results and discussion

Milestone

1. Insight into physiological behaviour of

the grapevine and grapes during

ripening

2. Practical indicators of grape

composition parameters associated

with a particular wine style

3. Ability to produce different wine styles

from a particular vineyard

4. Collaboration with foreign universities

and institutions

5. Training of students and personnel

6. Publications and papers delivered

Achievement

A deeper understanding of the relationship

between the soil, vine behaviour and primary

and secondary metabolism, with specific

reference to the berries and expected wine

quality

Clear indicators found

Clearly shown

Well established and still continuing with

France, Italy (mainland and Sicily), Spain

Local (MSc) and foreign students (PhD)

trained

Numerous

Research was done on a Shiraz/R99 vineyard in the Stellenbosch region with the purpose ofdefining environmental, canopy and grape parameters that may be suitable as eventualpractical indicators for obtaining particular styles of grapes and wine.

Final report 5

During the course of the project, several facets of this aspect of cultivation were investigatedin collaboration with South African, Spanish, French, Italian and Swiss researchers. OnePhD study (University of Palermo, Italy) and one MSc study (University of Stellenbosch,South Africa) during two growth seasons (2002/03 - 2003/04) were also completed as part ofthis project. Here, the publication of only the main focus of the project is presented. This isfollowed by some of the summaries (full length articles available as per list of publications) ofthe rest of the facets investigated as part of the project. Both the complete PhD dissertationand MSc thesis are available.

I. Role of Harvesting Time/Optimal Ripeness in Zone/Terroir Expression

JJ Hunter1, A Pisciotta2, CG Volschenk1, E Archer3 , V Novello4, E Kraeva5, A Deloire5, MNadal6

1)ARC Infruitec-Nietvoorbij, Private Bag X5026, 7599 Stellenbosch, South Africa.2)Dipartimento di Colture Arboree, Universita degli Studi di Palermo, Viale delle Scienze 11,90128 Palermo, Sicily, Italy3)Lusan Premium Wines, PO Box 104, 7599 Stellenbosch.4)Dipartimento di Colture Arboree, Via Leonardo da Vinci 44, I 10095 Grugliasco (TO), Italy5)Agro Montpellier, UMR 1083 « Sciences pour I'cenologie et la Viticulture », 2 place Viala,34060 Montpellier cedex 1, France.6)Departament de Bioquimica i Biotecnologia, Facultat d'Enologia de Tarragona, Ramon yCajal 70, 43003 Tarragona.kobush@infruit.agric.za

Acknowledgements : G.W. Fouche, L.F. Adams, J. Smith, W.J. Hendricks, C. Benn andPersonnel of the Nietvoorbij Experiment Farm for technical assistance. R. Chauhan for aminoacid analyses. The SA Vine and Wine Industry (through Winetech) for financial support. A.Louw for winemaking and F. Calitz for statistical analyses.

Key words : Grapevine, Shiraz, physiology, grape composition, ripeness level, wine quality,wine style.

AbstractOptimal ripeness is defined according to the style of wine that is required. The latter isultimately dictated by the market. Soil and climate may have a dictating effect on typicalexpression of wine. The level of grape and wine quality achieved and the potential forobtaining different styles of wine are determined by the integrated effect of the naturalcharacteristics of the terroir and technological intervention (long and short term cultivationpractices). The growth conditions that the grapevine is subjected to should allow optimalmetabolic activity in roots, permanent structure, canopy and grapes and the potential for theseorgans to develop and support each other until the desired grape quality and style is reached.Monitoring of morphological and physiological parameters in the canopy and grapes, ultimatelydisplaying the integrated effect of the growth environment, is critical in our quest for findingindicators that may be associated with a particular grape and wine style. This has not beensystematically investigated.

Results of collaborative research done on a Shiraz/R99 vineyard in the Stellenbosch region,South Africa, with the purpose of defining environmental, canopy and grape parameters thatmay be suitable as eventual practical indicators for obtaining particular styles of grapes andwine, are presented. Vines were vertically trellised and spaced 2.75 x 1.5 m in north-southorientated rows on a Glenrosa soil and a west-facing slope. Microsprinkler-irrigation wasapplied at pea berry size and at veraison stages. The 1.4 m canopies were shoot-positionedand topped. Fortnightly sampling was done from berry set up to two weeks post-veraison,after which harvesting for wine making was done approximately every four days. Microclimate,

Final report 6

vegetative, reproductive and physiological parameters were investigated and changes duringalcoholic fermentation monitored at each harvesting stage. Wines were made and analysed.Similarities in patterns as well as various ratios between the different parameters wereinvestigated. Results are argued against canopy performance, carbon allocation, waterrelations, production level, and sugar, acidity, anthocyanin, phenolic and tannin contents of thegrapes as well as wine quality and composition. Ratios for potential practical use indetermining optimal grape quality, time of harvesting and expected wine style are discussed.

IntroductionWorld-wide, many styles of wine exist that are judged by the general consumer and the wineexpert, both of whom are very important in quality classification, marketing and selling of aparticular wine. Although the grape composition and winemaking procedure are eminentlydictating the final product, a particular style of wine should be directed by market requirements.Optimal ripeness for a particular style of wine is therefore also ultimately dictated by the marketand the result of a three way communication (grape producer, winemaker, market). The levelof grape and wine quality achieved, the expression of a typical terroir-related character in thewine and the potential for obtaining different styles of wine are determined by the integratedeffect of the natural characteristics of the chosen terroir (soil and climate) and technologicalintervention (long and short term cultivation practices) (Jackson & Lombard, 1993; Calo et al.,1996; Hunter & Archer, 2001a, 2001b; Deloire et al., 2002). This is applicable to a singlevineyard or different vineyards, the latter which may lead to viticultural zoning.

Monitoring of morphological and physiological parameters in the canopy and grapes,displaying an integrated effect of the growth environment (aboveground and subterranean),along with wine composition and sensorial quality, are critical aspects in our quest for findingindicators that may be associated with a particular grape and wine style. Despite the crucialimportance of the latter for viticulture and oenology and the essentiality of timely harvesting fora quality product, it has only been occasionally studied/reviewed (Gonzalez-San Jose et al.,1991; Lacey et al., 1991; Calo et al., 1996; Ribereau-Gayon et al., 2000; Bisson, 2001; Hunteret al., 2004). The accumulation patterns of secondary metabolites (e.g. phenolic compounds,flavour compounds, anthocyanins, proteins and glycosidically-bound secondary compounds) inmaturing berries received the most attention (e.g. Gholami et al., 1996; Downey et al., 2003;Hilbert ef al., 2003; Hunter et al., 2004). To our knowledge, a purposeful and systematicinvestigation over a wide range of ripeness levels, involving physiological and morphologicalmeasurements on the canopy and grapes as well as wine composition and quality, and withthe purpose of finding practical indicators of optimal harvesting points and different wine styles,has not been done before. Such a study is critical in order to, amongst others, fully expressthe potential of a particular terroir in the eventual wine quality, irrespective of wine stylerequired.

It is envisaged that the study will contribute to understanding the relationships between leaf(source) and grape (sink) seasonal metabolic changes and optimal grape and wine quality.The objectives of the study were to determine the importance of different seasonal growthphases for grape and wine quality and grape parameters and ratios of selected grape chemicalconstituents at different levels of ripeness that would indicate an optimal harvesting time aswell as different styles of wines. In this way, practical parameters that would enable producersto manipulate and judge optimal grape and wine quality associated with a particular style maybe found.

Materials and MethodsVineyard and viticultural practicesA seven-year-old Vitis vinifera L. cv. Shiraz (clone SH1A) vineyard, grafted onto Richter 99(Vitis Berlandieri x Vitis rupestris) (clone RY2A), was used. The vineyard is located on theExperiment Farm of ARC Infruitec-Nietvoorbij in Stellenbosch (Western Cape). The area isaffected by a Mediterranean climate. The vines are spaced 2.75 m x 1.5 m on a Glenrosa soil

Final report 7

(Soil Classification Working Group, 1991) with western aspect (26° slope) and trained onto a 7-wire (cordon wire and three sets of movable wires) Lengthened Perold (Vertical ShootPositioned) Trellising System (Zeeman, 1981). Vines were micro-sprinkler irrigated at pea sizeand at veraison stages (12 hours @ 32L/hour). Vines were pruned to two-bud spurs with aspur spacing of approximately 15 cm. Rye was used as cover crop during winter. Normalcultivation practices for the production of healthy grapes were used.

TreatmentsVines were all similarly pruned and the canopies shoot positioned and topped (Hunter, 2000).Shoots were vertically positioned in line with corresponding spurs. Topping (30 cm above thetop wire) was done twice during the period berry set to pea size and comprised the removal ofup to 30 cm of shoots. Ripeness level harvesting dates were replicated three times. Ten vinesper replicate were used until 14 days post-veraison and 15 vines per replicate there after. Intotal, sampling was done at 19 dates, including 16 harvesting dates for winemaking purposes(from 14 days post veraison).

Measurements and analysesCanopy dimensions and microclimate, soil water, canopy growth and physiology, and grapegrowth and composition were monitored from 14 days after berry set stage at two-weekintervals from separate groups of vines until 14 days post-veraison. During the last stages ofripening, parameters were monitored approx. every 4 days and wines were additionally madeof each harvest during this period. Seven shoots (with bunches) per replicate andsampling/harvesting date were randomly selected and used for all vegetative and reproductivedeterminations and chemical analyses. Wines were made from a minimum of 40 kg grapesper wine. Microclimatic parameters in the bunch and canopy were continuously recordedduring the whole grape monitoring period by means of probes and data loggers. Levels ofgrape constituents obtained during the first phase of the berry growth period (up to veraison)were correlated to levels obtained during the last stages of grape ripening. Grape parametersand ratios of the different components analysed were correlated with wine sensorial qualityresults in order to identify factors that would be best suited as indicators of harvesting points fordifferent styles of wine and which may be used to predict different wine sensorial qualitycharacteristics.

Photosynthetic activity of primary and secondary leaves close to the bunch zone wasmeasured during mid-morning (from 10:00) using an open system ADC portablephotosynthesis meter (The Analytical Development Co., Ltd., England). Leaf and stem waterpotential of primary and secondary leaves was determined during early afternoon using apressure chamber (Scholander et al., 1965). Primary and secondary shoot length, leaf massand area, bunch and berry mass and volume, and berry pedicel, brush and seed browningwere determined at all stages. Light intensity in the bunch zone of the canopy was measuredduring mid-morning by means of a LICOR Line Quantum Sensor. The canopy dimensions andmicroclimate were visually scored (Hunter, 1999 - based on that of Smart et al., 1990).

Total soluble solids (°B), total titratable acidity (as g/L tartaric acid), and pH were analysedaccording to standard methods. Malic and tartaric acid as well as sucrose and glucose inleaves (primary and secondary), whole berry, berry pulp and berry skins were extracted andanalysed by GLC according to a method described by Hunter & Ruffner (2001). Skinanthocyanins and total phenolics were extracted and determined according to Hunter et al.(1991). Whole fresh berry anthocyanins, tannins (skins and seeds), total phenolic index, colourintensity, and hue/tint were extracted and determined according to the methods described byRibereau-Gayon et al. (2000) and Cliff et al. (2002). Wines were analysed using the samemethods. Amino acids were analysed by gradient high performance liquid chromatography(Bidlingmeyer et al., 1984; Rautenbach, 1999). Free-amino-nitrogen (FAN) in the must wasdetermined according to an Auto Analyzer method using ammonium sulphate as reference(Anonymous, 1974).

Final report 8

Grapes of all harvests were cooled to the same temperature (20 °C) before processing. Wholeberries, skins, seeds, pomace and wine were analysed for each ripeness level. Grapes weredestemmed, crushed and the pomace inoculated with commercial yeast (VIN 13). Alcoholicfermentation took place at a controlled temperature of 24 °C (di-ammonium phosphate andSO2 were added). The skins were pushed through three times per day. Fermentation on theskins averaged five days, after which the pomace was pressed. Skins and pomace wereanalysed during fermentation on the first, second, fourth and fifth day after crushing (seedswere only analysed on the fifth day after crushing - data not shown in this paper). Evolution ofcolour density (A520 + 420) and total phenolic content (A28o) was monitored in the pomace andskins. Total flavan-3-ol (catechin tannin) (DMAC analysis) was determined in the seeds fromintact berries and after pressing. Wines were made similarly. The degree of alcohol,phenolics, colour intensity, and colour density were determined in the different wines. Wineswere made similarly and were organoleptically evaluated by a trained panel. In addition, wineswere analysed for anthocyanin, tannin, and phenolic contents (also directly after the skincontact period).

Statistical layout and analysesThe experiment was laid out as a randomised block design with three block replications. Thetreatment design was a 2 x 19 factorial with factors being two treatments [control and canopymanagement (not described)] and 19 harvesting times. Sensorial evaluations of wines weredone using four seven-member panels, each tasting all replications of the different ripenesslevels. Measurements were taken for three growth seasons. In this paper, data of the controltreatment for the last two growth seasons are presented. Trends of variables and ratios duringthe season for the control treatment were compared for two growth seasons and combined ifnecessary. Coefficients were compared using t-test. Moving averages were used when datawere presented on figures. Variables with significant trends were selected and correlated withwine sensorial evaluation quality variables. Stepwise regression was performed to select thebest subset of variables to predict different wine quality variables. Grape parameter and ratiogroups for the identification of distinct wine styles were compared using t-test. A significancelevel of 5 % and less was applied to all data.

Results and DiscussionThe growth conditions and basic canopy management (e.g. trellising, shoot positioning &topping) allowed the attainment of primary shoot lengths of 1.2 - 1.3 m, carrying 1 3 - 1 6leaves, which were almost perfectly inside the general criteria for obtainment of high qualitygrapes as suggested by Hunter (2000) and Nadal et al. (2001) [(basic canopy managementnormally shifts the canopy composition balance (mostly additional secondary shoot growth) toa more favourable situation in terms of canopy function, continued canopy support to grapes,and grape quality - see also Hunter (2000) and Hunter et al. (2004)]. Radiation interceptedabove the canopy and reflected from the soil into the canopy was highest in October (duringbud break) and lowest in March (during late harvesting) (data not shown). Despite that,canopy temperatures increased during this time until February and bunch temperatures untilMarch (February is normally the hottest month during the harvesting period in South Africa)(data not shown). Bunches were not directly sun-exposed, which is generally accepted asbeing preferable in terms of grape quality. Similar trends were found for vertically trainedcanopy management treated Chenin blanc vines (Volschenk & Hunter, 2001).

Primary shoot length already stabilized from around berry set/pea size when topping started(data not shown). In both growth seasons, primary leaf area showed a decline (senescence)from around 8 weeks after veraison, whereas secondary shoot leaf area was maintained forthe whole harvesting period (Fig. 1a & 1b). Secondary shoot leaves in the canopy generallyphotosynthesised at a much higher water use efficiency (WUE), thereby increasing WUE of thewhole canopy [see also results of Hunter & Visser (1988) and Hunter (2000) for CabernetSauvignon and Sauvignon blanc] (Fig. 2). Interestingly, WUE of vines peaked approximately 8

Final report 9

weeks after veraison for both primary and secondary leaves. Secondary leaves were newlymatured during the ripening period and therefore very efficient during this time. This is alsoevident from higher sucrose production of secondary shoot leaves for most of the season (Fig.3). Generally, glucose concentrations of primary and secondary leaves were similar and malicacid concentrations of secondary leaves higher than those of primary leaves (data not shown).Lateral shoots effectively sustained canopy sufficiency and led to a higher canopy quality whichwould provide a buffer capacity against terroir-related stress conditions and facilitate continuedcontribution to grape composition as well as the permanent structure of the vine for a longerperiod of time (see also Hunter, 2000). On terroirs with, e.g. low soil water holding capacityand low temperatures during the later harvesting period, and which induce early leafabscission, lateral shoots would make a significant contribution to grape ripening. Sucroseproduction by both primary and secondary leaves became restricted during the last harvestingstages (from approximately 8 weeks after veraison). Thus, although leaf area (particularlysecondary leaf area) was still largely intact, this would seem to indicate a physiological endpoint as regards maximum leaf function and support. A further increase in soluble solidcontent was most probably increasingly dependent on secondary shoot leaves, the reservepool of the vine and redistribution of carbon.

Bunch mass and volume (data not shown) as well as berry mass and volume (Fig. 4) reachedhighest values about 3 weeks post veraison, after which mass and volume of berries werereduced by approximately 40 % until the final harvesting date. Except for physiologicalimplications, this also has serious financial repercussions. If a reduction in berry mass (alongwith other parameters) is to be used as a quality criterium for red cultivars, max./min. sizesshould be established for different cultivars and situations along with concomitant qualitychanges. Payments should be adjusted accordingly. The bunch and berry mass and volumepeaks corresponded with the first soluble solid peak (Fig. 5) and also coincided with the firstpeak sucrose, glucose, tartaric and malic acid concentrations in both primary and secondaryleaves (data not shown). This illustrates the fine balance between vegetative and reproductivegrowth, irrespective of the environmental conditions to which the vine is subjected. This findinghas significant implications for grape growing on different terroirs. If leaf production is limited,replenishment of the berry with water and concomitant respiratory substrate (primarily sucrose)would also decrease. In such an event, the berry transpires more water than it can gain fromwater potential and pressure flow gradients, leading eventually to a metabolic deceleration,berry shrivelling and eventual decay (Greenspan et a/., 1994; McCarthy & Coombe, 1999;Hunter & Ruffner, 2001). This will also be affected by the extent to which the attraction(metabolic activity) of grapes for supply from the leaves is promoted by bunch microclimate. Inagreement with these arguments, the must soluble solid content showed a first peak around 3weeks after veraison, where after further increases, most probably primarily as a result ofconcentration, occurred. Similar patterns were noticeable for total titratable acidity(decreasing) and pH (increasing). Seasonal sucrose concentration patterns in the berryshowed a remarkable correspondence with soluble solid content, the initial peakconcentrations in the whole berry, pulp and skin occurring from 2 weeks (skin) to 3 weeks(berry) after veraison; glucose concentrations first peaked about 1 week earlier (Fig. 6a & 6b).Sucrose concentrations in the berry seemed to increase during the last stages of ripening,most probably because of concentration together with restricted hydrolysis. This correspondedwith a decrease in sucrose concentration in primary and secondary leaves (Fig. 3). Despite anincrease in plant leaf and stem water potential (data not shown) during this time, berry size wasreduced (Fig. 4); support from the leaves to bunches (phloem transport) apparently becamerestricted and berries seemed to function more independently. Berry contents seemed tobecome more and more dependent on physical changes and progressive dehydration (seealso Dreier et a/., 1998; Dreier et a/., 2000). In general, critical physiological changes in theleaves and berries seemed to occur at approximately 3 weeks and 8 weeks after veraison.Amino acid and total nitrogen patterns in primary leaves, berries and skins showed that prolinein particular increased noticeably from veraison to the last harvesting stage (data not shown),indicating stress/senescence. This is in agreement with results found by Hilbert et a/. (2003).

Final report 10

The rest of the amino acids as well as total nitrogen increased in the primary leaves (nitrogenreserve accumulation), decreased in the whole berry (despite the berry size reduction) andmostly kept stable in the skins during the later stages of ripening, clearly indicating reducedtransport to the berry pulp (data not shown). Since the presence of amino acids is generallyconsidered to be favourable to fermentation and flavour development in wine (Rapp & Versini,1996), this would contribute to reduced berry quality during the later ripening stages.

Berry skin anthocyanin and phenolic patterns (based on dry mass extraction) correspondedwith soluble solid and hexose patterns of the berry, peaking early during ripening (3 weeksafter veraison) with a slight increasing tendency there after up to approximately 8 weeks afterveraison (Fig. 7). The anthocyanin released by whole, fresh, crushed berries measured at pH3.2 [approximately equal to extraction during fermentation (Ribereau-Gayon et a/., 2000)]essentially followed similar patterns than those of skin phenolic compounds, peakingapproximately 8 weeks after veraison and keeping virtually stable after that (Fig. 8). Incontrast, from approximately 8 weeks after veraison, the contribution of seed-tannin increasedfor a further 2 weeks where after it stabilised, whereas total phenolics and tannin continued toincrease until the final harvesting date (similar to berry soluble solids and pH). The flavan-3-olmonomer (catechin) content was also higher during the last harvesting stages. [The differencein anthocyanin curves obtained at pH 3.2 versus that obtained at pH 1.0 (at which pHproteophospholipid tonoplast membranes of cells are ruptured, protein bonds broken, contentsof vacuoles released and all anthocyanins extractable and solubilised in solution) showed thatanthocyanins were not effectively released during extraction at pH 3.2 in spite of their potentialavailability (data not shown)]. At the same time, the pomace:juice ratio increased, indicatinglesser and lesser sap recovery from grapes as ripening proceeded (data not shown). This wasalso evident from the decrease in berry volume (Fig. 4). Extraction from whole berries waspositively affected by the reduction in berry size as ripening continued. Similar trends werefound when berries were tasted (data not shown). Although the colour of the rachis, andpedicel and brush of the berry did not change much during the season, seeds changed from agreen-brown to white-brown colour and finally to a brown colour.

The extraction of anthocyanin, phenolics and tannin into the wine during fermentationincreased the longer the ripening period (Fig. 9a - 9c). However, during the later stages ofripening, extraction seemed to be completed. Both anthocyanins and phenolics in the wineincreased with fermentation up to the fourth day. The tannin content of the wine decreasedduring fermentation. Longer fermentation (up to 20 days) made no further contribution (datanot shown).

In the wine, anthocyanin, tannin, phenolic and total flavan-3-ol monomer (catechin) contents aswell as hue/tint and redness followed similar patterns to those found in grapes (Fig. 10).Further, sensorial quality of the wine also corresponded with the above findings on the canopyand grapes, clearly showing the emergence of different wine styles as grape ripeningproceeded (Fig. 11). The probable alcohol content and residual sugar of the wine continued toincrease throughout the harvesting period. Since the seasons differed climatically, the finalharvesting dates and number of harvesting dates (11 and 16) for winemaking were alsodifferent for the two seasons. However, for both seasons, after approximately 2 months afterveraison, there seemed to be no further improvement in any wine quality parameter, includingtotal wine quality impression. In fact, the acceptability of the acidity as well as aroma intensitydecreased during later ripening stages. Five weeks to approximately 8 weeks after veraisonseemed to be a period during which increasingly more concentrated wines with relatively loweventual wine probable alcohol content were obtained. After this period, no furtherimprovement in grape and wine quality was obtained from leaving the grapes on the vine. Infact, wines obtained additional jammy flavours.

It is hypothesised that basic metabolic changes would be controlled by homeostaticmechanisms. Various ratios of leaf, berry and wine chemical constituents at the different

Final report 11

harvesting dates (ripeness levels) were therefore calculated with the purpose of findingindicators of optimal harvesting time as well as different styles of wines, with consideration ofthe aforementioned trends. In this way, practical parameters that would enable producers tomanipulate and judge optimal grape and wine quality associated with a particular style areenvisaged. The ratios and other grape parameters were statistically reduced to those showingthe best correlations with wine sensorial quality parameters. The eventual grape parametersand ratios and correlations with the different wine sensorial quality parameters are presented inTable 1. The ratios were further subjected to a stepwise regression analysis, showing thegrape parameters and ratios that can best be used to predict the different wine sensorialquality parameters (Table 2).

Stepwise regression was also used to identify parameters that would best indicate differentwine styles, the result being ratios of berry mass:tannin(whole berry extract), °B:titratableacidity, pH:titratable acidity and °B:tannin(whole berry extract). However, in order to indicatethe different wine styles in terms of practically applicable parameters, commonly used grapeparameters that are analysed with standard methods were used with consideration of all theratios and grape parameters showing correlation and prediction value with the different winesensorial quality parameters. These parameters, i.e. °B, titratable acidity, pH, pH:titratableacidity, and °B:titratable acidity, showed highly significant correlation with the parametersselected by stepwise regression. Their values are indicated along with the wine sensorialquality parameters in Fig. 12. Despite the different climatic conditions of the two years, valuesof the ratios were well repeatable. [The values of other more complicated ratios based oncolour extraction and whole berry extraction (data not shown) would depend on the methodsused during analyses]. The grouping of the ratios corresponded to the observed growth andphysiological trends, indicating changes in source:sink relationships and interaction andreadiness of the grapes for distinct wine styles.

Different terroirs may change the criteria for optimal ripeness via an advancement orretardation and even inhibition of specific processes and their products (Hunter & Bonnardot,2002). This may lead to shorter harvesting periods (and thus a reduction in the options fordifferent styles of wines) and an alteration in the ratios of the different berry components,particularly when excessive stress-inducing conditions are experienced by the vine. Thegrowth conditions that the grapevine is subjected to should allow optimal metabolic activity inroots, permanent structure, canopy and grapes and the potential for these organs to developand support each other until the desired grape quality and style is achieved. The better suitedthe terroir and the concomitant management, the higher the potential for obtaining higher grapeand wine quality and different wine styles within a longer harvesting period. The pre-veraisonperiod is equally important in creating improved metabolic functioning of the leaves andgrapes; events during this period may be critical in the determination of eventual grape andwine quality (Carbonneau & Deloire, 2001; Hunter & Archer, 2001a, 2002; Ojeda et a/., 2002;Hunter et a/., 2004).

ConclusionsIdeally, the vineyard growth conditions should allow the prediction and distinction of differentharvesting times, representing different styles of wine that can then either be made unilaterallyor multilaterally by means of blending. Irrespective of the style of wine required, it is importantthat grapes are always harvested in a physicochemical state that would guarantee maximum(optimal) quality of the product for which they are intended. Cultivation practices should beapplied in such a way that the full grape quality potential of the vine is expressed on theparticular terroir. If the terroir and cultivation practices do not allow normal physiologicalperformance and grape potential to be fully expressed, as is often the case in practice, winequality and style options will be reduced. Judicious selection of terroir and long-term practicesis therefore critical.

Final report 12

The study clearly showed that optimal ripeness can not simply be described by the maximumaccumulation of grape components, but rather represents a complex, particular combination ofphysiological/biochemical changes (in leaves and grapes in particular growth phases), physicalchanges (homogeneity of vine and canopy structure as well as bunch and berry size), andrequired grape/wine style and market preferences.

Optimal harvesting time for a particular style of wine could be illustrated by using classicparameters of which information can be easily obtained during the growth season in thevineyard and in the winery by producers and winemakers. The global applicability of the winequality and style indicators found in this study (grape parameters and ratios) is currently furtherinvestigated. An extensive study on environmental impact (particularly soil water status) ongrape optimal ripeness is also underway.

Literature citedAnonymous, 1974. Technicon International Division SA. Operating manual for the TechniconNC-2 and NC-2P chromatography systems no. 9.Bidlingmeyer, B., Cohen, S.A. & Tarvin, T.L., 1984. Rapid analysis of amino acids using pre-column derivatisation. J. Chrom. 336, 93-104.Bisson, L, 2001. In search of optimal grape maturity. Practical Winery and Vineyard.July/August 2001, 32-43.Calo, A., Tomasi, D., Crespan, M. & Costacurta, A., 1996. Relationship betweenenvironmental factors and the dynamics of growth and composition of the grapevine. Proc.Workshop Strategies to Optimize Wine Grape Quality. Acta Hort. 427, 217-231.Carbonneau, A. & Deloire, A., 2001. Plant organization based on source-sink relationships:New findings on developmental, biochemical and molecular responses to environment. In:Roubelakis-Angelakis, K.A. (ed.), Molecular Biology & Biotechnology of the Grapevine, pp.263-280.Cliff, M.A., Brau, N., King, M.C. & Mazza, G., 2002. Development of predictive models forastringency from anthocyanin, phenolic and color analyses of British Columbia red wines. J.Int. Sci. VigneVin 36, 21-30.Deloire, A., Lopez, F. & Carbonneau, A., 2002. Reponses de la vigne et terroir. Elementspour une methode d'etude. Progres Agricole et Viticole 119, 78-86.Downey, M.O., Harvey, J.S. & Robinson, S.P., 2003. Synthesis of flavonols and expression offlavonol synthase genes in the developing grape berries of Shiraz and Chardonnay (Vitisvinifera L.). Austr. J. Grape and Wine Research 9, 110-121.Dreier, L.P., Hunter, J.J. & Ruffner, H.P., 1998. Invertase activity, grape berry developmentand cell compartmentation. Plant Physiol. Biochem. 36, 865 - 872.Dreier, L.P., Stoll, G.S. & Ruffner, H.P., 2000. Berry ripening and evapotranspiration in Vitisvinifera L. Am. J. Enol. Vitic. 51, 340-346.Gholami, M., Coombe, B.G., Robinson, S.P. & Williams, P.J., 1996. Amounts of glycosides ingrapevine organs during berry development. Austr. J. Grape and Wine Research 2, 59-63.Gonzalez-San Jose, M.L., Barren, L.J.R., Junquera, B. & Robredo, L.M., 1991. Application ofprincipal component analysis to ripening indices for wine grapes. J. Food Composition andAnalysis 4, 245-255.Greenspan, M.D., Shackel, K.A. & Matthews, M.A., 1994. Developmental changes in thediurnal water budget of the grape berry exposed to water deficits. Plant, Cell and Environment17, 1-7.Hilbert, G., Soyer, J.P., Molot, C, Giraudon, J., Milin, S. & Gaudillere, J.P., 2003. Effects ofnitrogen supply on must quality and anthocyanin accumulation in berries of cv. Merlot. Vitis 42,69-76.Hunter, J.J., 1999. Present status and prospects of winegrape viticulture in South Africa -focus on canopy-related aspects/practices and relationships with grape and wine quality. In:Proc. 11th GESCO Meeting, June 1999, Marsala, Sicily, Italy, p. 70-85.Hunter, J.J., 2000. Implications of seasonal canopy management and growth compensation ingrapevine. S. Afr. J. Enol. Vitic. 21, 81-91.

Final report 13

Hunter, J.J. & Archer, E., 2001a. Short-term cultivation strategies to improve grape quality.Proc. VIIIth Viticulture and Enology Latin-American Congress, 12 - 16 November 2001,Montevideo, Uruguay. On cd.Hunter, J.J. & Archer, E., 2001b. Long-term cultivation strategies to improve grape quality.Proc. VIIIth Viticulture and Enology Latin-American Congress, 12 - 16 November 2001,Montevideo, Uruguay. On cd.Hunter, J.J. & Archer, E., 2002. Paper actual de la gestio del fullatge i perspectives futures(Status of grapevine canopy management and future prospects). ACE Rivista d'Enologia 19,5-11.Hunter, J.J. & Bonnardot, V., 2002. Climatic requirements for optimal physiological processes:A factor in viticultural zoning. In: Proc. IVth International Symposium on Viticultural Zoning,17-20 June 2002, Avignon, France. 9 pp.Hunter, J.J., De Villiers, O.T. & Watts, J.E., 1991. The effect of partial defoliation on qualitycharacteristics of Vitis vinifera L. cv. Cabernet Sauvignon grapes. II. Skin color, skin sugar,and wine quality. Am. J. Enol. Vitic. 42, 13-18.Hunter, J.J. & Ruffner, H.P., 2001. Assimilate transport in grapevines - effect of phloemdisruption. Aust. J. Grape and Wine Research 7, 118-126.Hunter, J.J. & Visser, J.H., 1988. The effect of partial defoliation, leaf position anddevelopmental stage of the vine on the photosynthetic activity of Vitis vinifera L. cv. CabernetSauvignon. S. Afr. J. Enol. Vitic. 9(2), 9-15.Hunter, J.J., Volschenk, C.G., Marais, J. & Fouche, G.W., 2004. Composition of Sauvignonblanc grapes as affected by pre-veraison canopy manipulation and ripeness level. S. Afr. J.Enol. Vitic. 25, 13-18.Jackson, D.I. & Lombard, P.B., 1993. Environmental and management practices affectinggrape composition and wine quality - a review. Am. J. Enol. Vitic. 44, 409-^430.Lacey, M.J., Allen, M.S., Harris, R.L.N. & Brown, W.V., 1991. Methoxypyrazines in Sauvignonblanc grapes and wines. Am. J. Enol. Vitic. 42, 103-108.McCarthy, M.G. & Coombe, B.G., 1999. Is weight loss in ripening grape berries cv. Shirazcaused by impeded phloem transport. Aust. J. Grape and Wine Research 5, 17-21.Nadal, M., Menchon, J., Mateu, R. & Porta, M., 2001. Influence de la hauteur de palissage surla qualite du raisin de cv. Cabernet Sauvignon en climat mediterraneen. In: Proc. 12th GESCOMeeting, 3 - 7 July 2001, Montpellier, France, p. 401^06.Ojeda, H., Andary, C, Kraeva, E., Carbonneau, A. & Deloire, A., 2002. Influence of pre- andpostveraison water deficit on synthesis and concentration of skin phenolic compounds duringberry growth of Vitis vinifera cv. Shiraz. Am. J. Enol. Vitic. 53, 261-267.Rapp, A. & Versini, G., 1996. Influence of nitrogen compounds in grapes on aromacompounds of wines. Vitic. Enol. Sci. 51, 193-203.Rautenbach, M., 1999. The synthesis and characterisation of analogues of the antimicrobialpeptide Iturin A2. Ph.D. thesis (Biochemistry), University of Stellenbosch, p 2.9-2.24.Ribereau-Gayon, Y., Glories, Y., Maujean, A. & Dubourdieu, D., 2000. Handbook of Enology,Volume 2: The chemistry of wine and stabilization and treatments. John Wiley & Sons Ltd.Scholander, P.F., Hammel, H.T., Bradstreet, E.D. & Hemmingsen, E.A., 1965. Sap pressure invascular plants. Science 148, 339-346.Smart, R.E., Dick, J.K., Gravett, I.M. & Fisher, B.M., 1990. Canopy management to improvegrape yield and wine quality - principles and practices. S. Afr. J. Enol. Vitic. 11,3-17.Soil Classification Working Group, 1991. Soil classification - A taxonomic system for SouthAfrica. Department of Agricultural Development: Memoirs on natural agricultural resources ofSouth Africa no. 15, Department of Agricultural Development, Pretoria, South Africa.Volschenk, C.G. & Hunter, J.J., 2001a. Effect of seasonal canopy management on theperformance of Chenin blanc/99 Richter grapevines. S. Afr. J. Enol. Vitic. 22, 36-40.Zeeman, A.S., 1981. Oplei. In: Burger, J. & Deist, J. (eds.). Wingerdbou in Suid-Afrika. ARCInfruitec-Nietvoorbij, Private Bag X5026, 7599 Stellenbosch, South Africa, pp. 185-201.

able 1. Correlations

Grape

Parameter/ratio

°B must

TA must

pH must

pH:TA

°B:TA

°B:pH

Anth

Phenolics

Anth:Berry mass

Anth:Berry vol.

TA:Anth

Malic:Sucrose

Tartaric:Glucose

Malic:Glucose

Berry mass:Seed T

Berry mass:Tannin

Berry vol.:Seed T

Berry vol.:Tannin

TA:Phenolics

TA:Seed T

TA:Tannin

pH:SeedT

pH:Tannin

°B:Tannin

Berry mass:Anth

Berry vol.:Anth

TA:Anth

of Shiraz/R99 grape parameters/ratios with wine sensorial quality

Total aroma

intensity

0.63

-0.74

0.75

0.64

0.61

0.50

0.76

0.73

ns

0.39

-0.80

ns

-0.55

ns

ns

ns

-0.56

ns

-0.53

-0.82

-0.67

-0.50

ns

ns

ns

ns

-0.69

Berry

Aroma

0.55

-0.60

0.63

0.58

0.56

0.44

0.59

0.59

ns

ns

-0.60

ns

-0.57

-0.45

0.62

ns

-0.63

ns

-0.47

-0.73

-0.54

-0.62

-0.38

ns

ns

ns

-0.50

Acceptability

of acidity

Body

BERRIES

0.46

-0.40

0.40

ns

ns

0.48

SKIN

0.66

0.56

ns

ns

-0.52

ns

-0.40

ns

0.63

-0.68

0.74

0.64

0.62

0.49

I

0.67

0.64

ns

ns

-0.71

ns

-0.67

-0.48

Colour

0.71

-0.71

0.74

0.69

0.69

0.62

0.57

0.57

0.48

0.51

-0.70

-0.48

-0.54

-0.48

WHOLE BERRY EXTRACTION

0.52

ns

ns

ns

ns

-0.61

-0.46

ns

ns

ns

ns

ns

-0.49

0.54

ns

-0.57

ns

-0.53

-0.77

-0.61

-0.53

ns

ns

ns

ns

-0.61

0.63

-0.47

-0.67

-0.53

-0.59

-0.79

-0.70

-0.65

-0.55

-0.42

-0.46

-0.47

-0.66

Final report 14

variables.

Total quality

Impression

0.63

-0.72

0.70

0.63

0.61

0.53

0.64

0.63

ns

ns

-0.75

-0.39

-0.57

-0.45

0.50

ns

-0.58

ns

-0.56

-0.83

-0.67

-0.54

-0.42

ns

ns

Ns

-0.68

°B = °Balling, TA = Titratable acid, Anth = Anthocyanin

All correlations are significant at 5 % or less; ns = non-significant

Final report 15

Table 2. Stepwise regression analysis of Shiraz/R99 grape parameters/ratios with wine sensonal qualityvariables.

Variable entered

TA:Anth (skin)

Berry mass:Anth (whole berry extract)

Berry volume:Anth (whole berry extract)

Anth (skin)

pH must

Berry mass

Anth (skin)

°B:pH

pH must

Berry mass

Berry volume:Anth (whole berry extract)

Berry pH

TA:Anth (skin)

Anth (skin):Berry mass

Dependent variable/R

TOTAL AROMA INTENSITY

0.65

0.72

0.78

0.81

BERRY AROMA

0.39

0.47

ACCEPTABILITY OF ACIDITY

0.44

0.53

BODY

0.55

0.63

0.70

COLOUR

0.52

TOTAL QUALITY IMPRESSION

0.56

0.62

°B = °Balling, TA = Titratable acid, Anth = Anthocyanin

All correlations are significant at 5 % or less

•2002/03 2003/04

80 90 100 110 120 130 140 150 160 170 180 190 200

Days after budbreak

Fig 1 a. Seasonal variation in primary shoot leaf area.

-2002/03 2003/04

80 90 100 110 120 130 140 150 160 170 180 190 200

Days after budbreak

Fig 1b. Seasonal variation in secondary shoot leaf area.

Final report 16

5.0

4.5

4.0

3 5

3.0

2.5

2.0

1 5

1.0

0.5

0.0

Primary leavesSecondary leaves

90 100 110 120 130 140 150 160 170 180 190 200

Days after budbreak

Fig 2. Seasonal variation in water use efficiency of primary andsecondary leaves and the effect of berry ripeness level.

2.01.8 •

1.6-

1.4

1.2 •

1.0 •

0.8 •

0 .6 -

0 .4 -

0.2 -

0 . 0 -

Berry mass - - - - Berry volume

14

Primary leavesSecondary leaves

90 100 110 120 130 140 160 160 170 180 190 200

Days after budbreak

Fig 3. Seasonal variation in sucrose content of primary andsecondary leaves and the effect of berry ripeness level.

pH• Must soluble solids

70 80 90 100 110 120 130 140 150 160 170 180 190 200

Days after budbreak

Fig 4. Seasonal variation in berry mass and volume and the Figeffect of berry ripeness level.

32

130 140 150 160 170 180 190 200

Days after budbreak

5. Must soluble solids, titratable acid, pH and the effectof berry ripeness level.

Skin

70 80 90 100 110 120 130 140 150 160 170 180 190 200

Days after budbreak

40

Whole berry

90 100 110 120 130 140 150 160 170 180 190 200

Days after budbreak

Fig 6a. Seasonal variation in whole berry, pulp and skin sucrose Fig 6b. Seasonal variation in whole berry, pulp and skincontents and the effect of berry ripeness level. glucose contents and the effect of berry ripeness level.

Final report 17

0.0

Anthocyanin (A520)

Phenolics (A280)

1.0

0.9

0.6

07

06

05

0

0.3

02

0

100 110 120 130 140 150 160 170 180 190 200

Days after budbreak

Fig 7. Skin anthocyanin and phenolic content and the effect ofberry ripeness level.

Phenolics (a.u.)-Antho (mg/l)•Tannin (g/l)•Catechin (a.u.)

120 130 140 150 160 170 180 190 2

Days after budbreak

Fig 8. Whole berry phenolic content and the effect ofberry ripeness level.

- - - - -

110 130 150 170 190

Days after budbreak

Fig 9a. Extraction of anthocyanin into wine during fermentationas affected by berry ripeness level and fermentation duration.

•Day 1

•Day 2

Day 4

•After pressing

110 130 150 170 190

Days after budbreak

Fig 9b. Extraction of phenolics into wine during fermenta-tion as affected by berry ripeness level and fermentationduration.

-Day 1

•Day 2

Day 4

-After pressing

110 120 130 140 150 160 170 180 190 200

Days after budbreak

Fig 9c. Extraction of tannin into wine during fermentation asaffected by berry ripeness level and fermentation duration.

60

50

40

30

20

10-

0

Days after budbreak

— - - —Phenolics• Total phenolic index

TanninRedness

- - - Anthocyanin (/10)Seed tannin

Colour intensityHue/Tint

Fig 10. Wine phenolic content as affected by berry ripenesslevel.

Final report 18

90

80

70

GO

50

40

30

20-

10

0

120 130 140 150 160 170

Days after budbreak180 190

-Reducing sugarAcidity acceptability

-Body- Berry aroma

0.0

200

— - — Total aroma intensity— — — Total quality impression— - - —Alcohol

8 0 .

70

30

20

1 0 .

<19c>8.5a<3.3c<2.0e<0.40d

<21b>6b<3.4b<3.0d<0.50c

<23a>5.5c<3.6ab<4.0c<0.65b

25a5c3.8a5.5b0.80b

<5c>3.8a>5.5a>0.80a

- 0 5

-0 .0

135 140 145 150 155 160 165 170 1

-Reducing sugar• Acidity acceptability-Body• Berry aroma

Fig 11 . Wine sensorial quality, and alcohol and reducing sugar Fig 12. Grouping of different wine styles according to easilycontents as affected by berry ripeness level. measurable grape parameters and ratios.

Phenolic extraction during fermentation as affected by ripeness level of Syrah/R99grapes

M. Nadal1, N. Volschenk2 and J.J. Hunter2

1Departament de Bioquimica i Biotecnologia, Facultat d'Enologia de Tarragona, CerTA,Ramon y Cajal 70, 43003 Tarragona, mnrj@fe.urv.es2ARC Infruitec-Nietvoorbij, Private Bag X5026, 7599 Stellenbosch, South Africa.

Acknowledgements: G.W. Fouche, L.F. Adams, J. Smith, W.J. Hendricks, C. Benn, D.Wenn, E. Marais and Personnel of the Nietvoorbij Experiment Farm for technical assistanceand the SA Vine and Wine Industry (through Winetech) for financial support. Adele Louw forwinemaking.

Key words: Grapevine, Shiraz, fermentation, phenols, ripeness level, skins, seeds, wine

AbstractPhenolic (tannin and anthocyanin) extraction during fermentation of Syrah grapes wasinvestigated as part of an elaborate study to determine parameters that would indicate highgrape quality and different grape and wine styles. A Syrah/R99 vineyard, situated in theStellenbosch region (South Africa), was used. Vines are vertically trained and spaced 2.75 x1.5 m in north-south orientated rows on a Glenrosa soil and a west-facing slope.Microsprinkler-irrigation was applied at pea berry size and at veraison stages. Canopieswere suckered, shoot-positioned and topped, whereas leaves were removed at two stages.Fortnightly sampling was done from berry set up to two weeks post-veraison, after whichgrapes were harvested for analyses and winemaking approximately every four days. Sixwines were made per ripeness level.

Results obtained during the ripening period of the 2002/2003 growth season (from 17February to 24 March) are reported. Whole berries, skins, seeds, pomace and wine wereanalysed for each ripeness level. Grapes of all harvests were cooled to the sametemperature (20 °C) before processing. Grapes were destemmed, crushed and the pomace

Final report 19

inoculated with commercial yeast (VIN 13). Alcoholic fermentation took place at a controlledtemperature of 24 °C (di-ammonium phosphate and SO2 were added). The skins werepushed through three times per day. Fermentation on the skins averaged five days, afterwhich the pomace was pressed. Skins and juice were analysed on the first, second andfourth day during fermentation. On the fifth day after crushing (at pressing), skins, juice andseeds were analysed. Total soluble solids, titratable acidity, pH, anthocyanins, tannins andphenolics were analysed in the whole berries. Evolution of colour density (A520 + 42o) and totalphenolic content (absorbance at 280nm) was monitored in the pomace and skins.Proanthocyanidin content (DMAC analysis) was determined in the seeds from intact berriesand in the seeds after pressing. The degree of alcohol, phenolics, colour intensity and colourdensity were determined in the different wines.

The °Balling of the berries reached a high at approximately 11 March (178 days after budburst). This pattern was similar to that of the anthocyanin, tannin and total phenolic contentsof the berry (whole berry extraction) and coincided with the reduction in berry size due towater loss. After 11 March extraction of the different phenolic compounds seemed not to beaffected by the decrease in berry size. From 17 March no further extraction from the skins(skin extraction after 5 days of fermentation) occurred, hence the stable colour density andtotal phenolic patterns of the skins during this period. The colour density and total phenoliccontent of the skins during fermentation showed a clear distinction between harvest dateswith higher extraction occurring from 11 March to the last harvest date, resulting in lowremaining values in the skins after five days of fermentation. The proanthocyanidin contentof the seeds only slightly decreased during the course of ripening. However, the seeds wereheavily depleted during fermentation of the harvests following that at approximately 6 weeksafter veraison, a trend which is completely opposite to the sugar content of the berries. Thecolour density and total phenolic content of the wine followed similar patterns to those of theberries.

Sugar Loading and Phenolic Accumulation as Affected by Ripeness Level ofSyrah/R99 Grapes

A. Deloire1, E. Kraeva1, M. Martin2 et J.J. Hunter3

1Agro Montpellier, UMR 1083 « sciences pour I'cenologie et la viticulture », 2 place Viala,34060 Montpellier cedex 1, France.2INRA, Unite experimentale de Pech Rouge, 11430 Gruissan, France.3ARC Infruitec-Nietvoorbij, Private Bag X5026, 7599 Stellenbosch, South Africa.deloire@ensam.inra.fr; hunterk@infruit.agric.za

Remerciements: Les auteurs remercient le personnel de ARC Infruitec-Nietvoorbij (TheFruit, Vine and Wine Institute of the Agricultural Research Council; Stellenbosch, Afrique duSud) pour I'aide technique au vignoble et au laboratoire et Winetech .pour le financementdes recherches. Us remercient egalement I'Agro Montpellier et I'LJE experimentale de Pech -Rouge (INRA Narbonne, France) pour I'aide technique et la mise a disposition de personnel.

Key words: Vitis vinifera, bunch, sugar, microclimate, indicator, vine functioning.

AbstractSugar loading and phenolic accumulation in Syrah grapes were investigated as part of anelaborate study to determine parameters that would indicate high grape quality and differentgrape and wine styles on a particular terroir. The relationship between the dynamics of sugarloading and phenolic accumulation in the berries of a Syrah/R99 vineyard, situated at theARC Infruitec-Nietvoorbij , in the Stellenbosch region (South Africa), was investigated frompea size stage (green berry) to late maturity. Vines were vertically trained and spaced 2.75 x1.5 m in north-south orientated rows on a terroir with Glenrosa soil and a west-facing slope.

Final report 20

Microsprinkler-irrigation was applied at pea berry size and at veraison stages. The 1.4 m highcanopies were suckered, shoot-positioned and topped and accommodated by means ofthree sets of double wires. The dynamics of berry sugar loading were studied by a methodfrom Deloire et al, 2004 (under publication), the berry phenolic composition (total tannins andpolymerisation, proanthocyanidins, anthocyanins) was analysed by spectrophotometry andanthocyanins by HPLC. Sugar was used as physiological indicator of the plant-berry (source-sink) relationship and as bunch microclimatic indicator.

The total tannin (TT) component in the berry was synthesised from anthesis to veraison. TheTT concentration increased during the green berry growth stages and decreased duringripening as the berry increased in volume. The TT per berry also increased during the greenberry growth stages, but kept stable during ripening. When sugar content per berry is usedas physiological indicator, it is clear that anthocyanin biosynthesis occurred until a specificberry sugar content, i.e. 20 - 21 °Brix, is reached. After this point, anthocyanin evolution perberry seemed independent of berry sugar evolution, which is at that time mainly due toconcentration (berry water loss) than to loading. Thus, although berry sugar loading isdependent on photosynthetic activity of the leaves, the regulation of sugar phloem unloadingin the berry sink seemed to be, in part, affected by the microclimate that the berryexperienced. Berry sugar loading was not directly correlated with berry volume.

INDEX OF PhD DISSERTATION OF A. PISCIOTTA [University of Palermo, Italy, 2004- the full dissertation is available (in Italian)]

Introduction 11. Translocation 52. Source-Sink ratio 103. Source-Sink competition in different phenological stages 16

4. Vegetative and reproductive Sink strength modification 184.1. Phototropic and Geotropic shoot orientation 184.1.1. Introduction 184.1.2. Trellis system roles • 204.1.3. Shoot orientation 224.1.4. Material and methods 244.1.5. Results and discussions 284.1.6. Conclusions 324.1.7. Tables and figures 334.2. Source strength variation according to internal vine microclimate 414.3. Defoliation 464.4. Vertical shoot positioning 544.5. Vertical shoot positioning: Effect on vegetative, reproductive, qualitative and physiologicalparameters on cv Merlot 56

4.5.1. Material and methods 574.5.2. Results and discussions 604.5.3. Conclusions 654.5.4. Tables and figures 674.6. Source strength variation according to leaf age 744.7. Source-Sink modification 764.8. Topping 784.9. Topping: effect on vegetative, reproductive, qualitative and physiological parameters

82

4.9.1. Material and methods 834.9.2. Results 86

4.9.3. Conclusions4.9.4. Tables4.10. (A) Shoot positioning and topping: Effect on vegetative, reproductive,

physiological parameters on cv Merlot

4.10.1. Introduction (A)4.10.2. Material and methods4.10.3. Results and discussions4.10.4. Technical considerations4.10.5. Conclusions4.10.6. Tables and figures4.11.1. Material and methods (B)4.11.2. Results4.11.3. Conclusions4.11.4. Tables and figures4.12. Shoot homogeneity: shoot vigour/grape quality relationship4.12.1 Target4.12.2. Material and methods4.12.3. Results (cv Merlot)4.12.4. Tables4.12.5. Results (cv Cabernet sauvignon)4.12.6. Conclusions4.12.7. Tables5. General conclusions

Literature cited

Final report 21

9193

qualitative and99

99101105107108109121121126127136138139143146152155157163165

Shoot Positioning: Effect on Physiological, Vegetative and Reproductive Parameters(Part of PhD dissertation)

A. Pisciotta1, R. Di Lorenzo1 M.G.Barbagallo1, C.G. Volschenk2 & J.J. Hunter2

1 Dipartimento di Colture Arboree, Universita degli Studi di Palermo, Viale delle Scienze 11,90128 - Palermo, Sicily, Italy. antoninopisciotta(a)virqilio.it, rdiloren@unipa.it2 ARC Infruitec-Nietvoorbij, Private Bag X5026, 7599 Stellenbosch, South Africa.hunterk@arc.agric.za

AcknowledgementsLeonard Adams, Stawie Fouche, Fanie Fouche, Cindy Benn, Dolores Wenn, Elwina Marais,Willemien Hendricks and farm personnel of ARC Infruitec-Nietvoorbij for technicalassistance. The SA Vine and Wine Industry (through Winetech) and the University ofPalermo for financial support.

Key words: Merlot, shoot positioning, vegetative growth, reproductive growth, photosynthesis,water potential, light interception, grape composition.

AbstractThe effect of vertical shoot positioning and topping at different times during two growthseasons (2002/03 and 2003/04) on physiological, vegetative and reproductive parameterswas investigated in a vertically trellised Meriot/R99 vineyard located in the Stellenbosch area.Vines were spaced 2.7 x 1.5 m in north-south orientated rows. Micro-sprinkler irrigation wasapplied at pea size berry and at veraison stages. Shoots were positioned at berry set, peasize, veraison and post-veraison stages (3 weeks after veraison). After being positioned,they were immediately topped. Before positioning the canopy was in a "natural" conditionwith shoots hanging freely. Soil water typically varied according to the progress in the

Final report 22

season and with soil depth, decreasing towards the end of the season and increasing withdepth. The primary shoot length of the positioned shoots was on average approximately 100- 115 cm, being restricted by the relatively low trellising system. Shoot positioning andtopping had no marked effect on the growth of secondary shoots, but they had a noticeableeffect on the position of secondary shoots along the length of the primary shoots. Pea-sizeshoot positioning induced slightly lower light conditions in the bunch zone, because of thelow position of secondary shoot development on primary shoots. In spite of this, pre-veraison shoot positioning treatments allowed good all-round light distribution, which wouldpromote uniform bunch ripening and grape quality. The basal and apical stem and leaf waterpotential and photosynthetic activity decreased during the season as the leaves aged andthe plants lost water. A significant correlation was found for apical leaves between stem andleaf water potential.

Earlier shoot positioning (up to veraison) significantly increased the °Balling level of the must.Early shoot positioning (up to veraison) increased malic acid and sucrose contents, whereastartaric acid contents were slightly reduced and glucose contents were higher in pea size andveraison treatments. No significant differences between treatments were found for must pH.The earlier shoots were positioned, the more water was lost by the skins, resulting in aconcentration of skin contents. Pre-veraison shoot positioning and topping improved thecolour of the skins.

No practical difficulty was experienced when shoots were positioned early in the season, i.e.at berry set and pea size stages, whereas at and after veraison proper vertical positioningwas primarily restricted by shoot lignification and the tightness of tendrils on the wires.Bunches were also very sensitive to damage, which led to bunch rot and a reduction in yield.These are important considerations in terroirs where timely management is difficult.

Phototropic and Geotropic Shoot Orientation: Effect on Physiological, Vegetative andReproductive Parameters (Part of PhD dissertation)

A. Pisciotta1, R. Di Lorenzo1 M.G.Barbagallo1, C.G. Volschenk2 & J.J. Hunter2

1 Dipartimento di Colture Arboree, Universita degli Studi di PalermoViale delle Scienze 11, 90128 - Palermo, Sicily, Italy. antoninopisciotta@virgilio.it,rdiloren@unipa.it2ARC Infruitec-Nietvoorbij, Private Bag X5026, 7599 Stellenbosch, South Africa.hunterk@arc.agric.za

AcknowledgementsLeonard Adams, Stawie Fouche, Fanie Fouche, Cindy Benn, Dolores Wenn, Elwina Marais,Willemien Hendricks and farm personnel of ARC Infruitec-Nietvoorbij for technicalassistance. The SA Vine and Wine Industry (through Winetech) for financial support.

Key words: Merlot, shoot orientation, vegetative growth, photosynthetic activity, water potential,light interception, grape composition.

AbstractThe effect of shoot orientation during two growth seasons (2002/2003 and 2003/2004) onphysiological, vegetative and reproductive parameters was investigated in the Stellenboscharea in a Merlot/R99 vineyard with a vertical trellising system. Vines were spaced 2.7 X 1.5m in north-south orientated rows. Micro-sprinkler irrigation was applied at pea size berry andat veraison stages. Observations were done on vines with a natural distribution andorientation of phototropically (upward) and geotropically (downward) shoots on the samecordon.

Final report 23

Soil water typically varied according to the progress in the season and with soil depth,decreasing towards the end of the season and increasing with depth. Geotropic orientationreduced the primary and lateral shoot length as well as the primary and secondary shoot leafarea. With phototropic shoot position, secondary shoots were more evenly distributed alongthe primary shoots. Basal and apical stem and leaf water potential was lower with geotropicorientation than with phototropic orientation. This was particularly pronounced during theripening period. In spite of this, basal and apical leaf photosynthetic activity of thephototropically orientated shoots was higher than that of the geotropically orientated shoots,most probably because of more favourable microclimatic conditions experienced by theformer. Bunch mass and volume and length of bunches were not significantly affected byshoot orientation. Phototropic orientation of shoots noticeably increased glucose and tartaricacid contents of the berries, whereas sucrose, malic acid and citric acid contents werevirtually unaffected. In phototropically orientated shoots, less water was lost by the skins,favouring skin colour intensity. The results have important implications for bunch and berrycomposition uniformity and for trellising system selection on different terroirs.

Shoot heterogeneity effects in a Shiraz/R99 vineyard (PART OF MSc THESIS,University of Stellenbosch, 2004 - the full thesis is available)

H Cloete1, E Archer2, V Novello3& JJ Hunter4

1 Department of Viticulture and Oenology, Private Bag X1, Matieland, University ofStellenbosch, 7602 Stellenbosch, South Africa. hcloete@sun.ac.za2 Lusan Premium Wines, PO Box 104, 7599 Stellenbosch. earcher@distell.co.za3Dipartimento di Colture Arboree, I 10095 Grugliasco, Italy, vittorino.novello@unito.it4 ARC Infruitec-Nietvoorbij, Private Bag X5026, 7599 Stellenbosch, South Africa.kobush@arc.agric.za

AcknowledgementsThe authors wish to thank the personnel of the viticulture and grapevine physiologylaboratory of the ARC Infruitec-Nietvoorbij, Stellenbosch, for technical assistance. We thankDr. Martin Kidd for his assistance in analysing the data.

Key words: Shoot heterogeneity, physiology, vegetative growth, reproductive growth, grapecomposition.

AbstractThe effect of shoot heterogeneity on vegetative and reproductive growth parameters, vinephysiology and grape composition was investigated in a Shiraz/Richter 99 vineyard.Comparisons between underdeveloped (typically shorter and less ripened at veraison) andnormally developed shoots in both shaded and well-exposed canopies were made.Compared to underdeveloped shoots, normal shoots had a larger total leaf area, due to thehigher occurrence of secondary shoots as well as larger leaves on primary and secondaryshoots. Since the physiological activity of the leaves from normal shoots was higher thanthat from underdeveloped shoots, higher levels of total carbohydrates were produced andstored in the former. Starch was more evenly distributed over the whole shoot length in thelonger and thicker normally developed shoots compared to the underdeveloped shoots. Thelarger clusters of the normally developed shoots were evidence of their more favourable totalleaf area per gram berry mass. Berries from the normally developed shoots were smaller atfive weeks after veraison than those from underdeveloped shoots, displaying a higher skin topulp ratio and therefore higher anthocyanin and total phenolic extraction potential forwinemaking. The peculiar absence of large differences in grape composition betweennormally and underdeveloped shoots indicated that assimilates needed for berry ripening of

Final report 24

the latter originated in organs other than the leaves [e.g. from adjacent normal shoots andthe rest of the permanent structure of the vine (cordon, trunk, roots)]. The larger differencesin berry size that occurred between shoot types in the shaded compared to the well-exposedcanopies may be evidence for this. The photosynthetic activity of shoots was lower inshaded than in exposed canopies. The total carbohydrate production of the normal shoots inshaded canopies seemed insufficient to supply in the ripening needs of their own clustersand of the shoot itself as well as the ripening of stem tissue and clusters of theunderdeveloped shoots in the canopy. This was illustrated by the lower levels of starch thataccumulated in the normal shoots from shaded compared to that of exposed canopies. Vineshoot heterogeneity clearly led to visible and physiological imbalances that would impactnegatively on grape and wine quality as well as production costs and should therefore beavoided on any terroir.

4. Accumulated outputs

Technology developedIndicators of grape and wine quality and style

Human resources developed/trainedFarm personnel and technicians in canopy management; 1 international PhD student; 1national MSc student; various producers

PatentsNone

Publications (popular, press releases, semi-scientific, scientific)Deloire, A., Ojeda, H., Wang, Z., Hunter, J.J., Paez, C, Martin, M. & Carbonneau, A., 2003.Grapevine water status and berry ripeness - consequence for vineyard cultural practices.Proc. 27th S.A. Society for Enology and Viticulture Conference, 5 - 7 November 2003,Somerset West, South Africa.

Pisciotta, A., 2004. Modificazione dei rapporti "source-sink" in Vitis vinifera L. mediantetecniche di gestione della chioma. PhD dissertation, University of Palermo, Italy.

Cloete, H., 2004. The effect of shoot heterogeneity on the physiology and grape compositionof Shiraz/Richter 99 grapevines. MSc thesis, University of Stellenbosch, South Africa.

Temperli, T., Louw, A. & Hunter, J.J., 2004. Weinbereitung am Kap der guten Hoffnung.Schweiz. Z. Obst-Weinbau 18, 10 - 13.

Hunter, J.J., 2004. Travel report of visit to Italy (Sicily). Collaborative research between the

University of Palermo, Sicily, and ARC Infruitec-Nietvoorbij, South Africa. 1 6 - 2 4 May 2004.

Hunter, J.J., 2004. Travel report of visit to France, Spain & Italy. Collaborative research

between AGRO Montpellier (France), University of Tarragona (Spain), University of Torino

(Italy) and ARC Infruitec-Nietvoorbij (South Africa). 7 - 2 2 August 2004.

Deloire A. & Hunter J.J., 2005. Berry composition as affected by bunch exposure. ProgresAgricole et Viticole 122, 151-157.

Hunter, J.J., Pisciotta, A., Voschenk, C.G., Archer, E., Novello, V., Deloire, A. & Nadal, M.,

Final report 25

2005. Role of harvesting time/optimal ripeness in zone/terroir expression. Proc. JointConference (SASEV, OIV, GESCO) on Viticultural Zoning, 1 5 - 1 9 November 2004, CapeTown, South Africa. In press.

Nadal, M., Volschenk, C.G. & Hunter, J.J., 2005. Phenolic extraction during fermentation asaffected by ripeness level of Syrah/R99 grapes. Proc. Joint Conference (SASEV, OIV,GESCO) on Viticultural Zoning, 15-19 November 2004, Cape Town, South Africa. In press.

Deloire, A. & Hunter, J.J., 2005. Sugar loading and phenolic accumulation as affected byripeness level of Syrah/R99 grapes. Proc. Joint Conference (SASEV, OIV, GESCO) onViticultural Zoning, 15 -19 November 2004, Cape Town, South Africa. In press.

Pisciotta, A., Di Lorenzo, R., Barbagallo, M.G., Volschenk, C.G. & Hunter, J.J., 2005.Phototropic and geotropic shoot orientation: Effect on physiological, vegetative andreproductive parameters. Proc. Joint Conference (SASEV, OIV, GESCO) on ViticulturalZoning, 15 -19 November 2004, Cape Town, South Africa. In press.

Pisciotta, A., Di Lorenzo, R., Barbagallo, M.G., Volschenk, C.G. & Hunter, J.J., 2005. Shootpositioning: Effect on physiological, vegetative and reproductive parameters. Proc. JointConference (SASEV, OIV, GESCO) on Viticultural Zoning, 1 5 - 1 9 November 2004, CapeTown, South Africa. In press.

Cloete, H., Archer, E. & Hunter, J.J., 2005. Shoot heterogeneity effects in a Shiraz/R99vineyard. Proc. Joint Conference (SASEV, OIV, GESCO) on Viticultural Zoning, 1 5 - 1 9November 2004, Cape Town, South Africa. In press.

Varvaro, G., Nadal, M., Lopez, N. & Hunter, J.J., 2005. Influencia de la maduracion y calidaddel vino tinto de Syrah. Proc. Gienol Symposium, 3 - 4 June, Palencia, Spain.

Presentations/papers deliveredOptimal ripeness. J.J. Hunter & E. Archer. SASEV Wine forum. 18 Jan 2002, Stellenbosch.

Seasonal canopy management : Effect on microclimate, red & white grape quality & winestyle. J.J. Hunter & E. Archer. Seminar to VinPro Consultants, 7 November 2002, Robertson.

Grape and wine phenolics. Seminar to VinPro Consultants, 7 November 2002, Robertson.

Grapevine water status and berry ripeness - consequence for vineyard cultural practices.Deloire, A., Ojeda, H., Wang, Z., Hunter, J.J., Paez, C, Martin, M. & Carbonneau, A., 2003.SASEV Congress, Nov. 2003, Somerset West.

Determining optimal ripeness. Hunter, J.J., Volschenk, C.G., Pisciotta, A., Archer, E., Nadal,M., Novello, V. & Deloire, A. SASEV-Congress, Nov. 2003, Somerset-West.

Shoot positioning: Effect on grape parameters. Pisciotta, A., Di Lorenzo, R. & Hunter, J.J.SASEV-Congress, Nov. 2003, Somerset-West.

Shoot positioning: Effect on physiological parameters. Pisciotta, A., Di Lorenzo, R. & Hunter,J.J. SASEV-Congress, Nov. 2003, Somerset-West.

Shoot heterogeneity effects in a Shiraz/R99 vineyard. Cloete, H., Archer, E. & Hunter, J.J.SASEV-Congress, Nov. 2003, Somerset-West.

Shoot heterogeneity effects in a Shiraz/R99 vineyard. Cloete, H., Archer, E. & Hunter, J.J.Joint Conference (SASEV, OIV, GESCO) on Viticultural Zoning, 1 5 - 1 9 November 2004,

Final report 26

Cape Town, South Africa.

Role of harvesting time/optimal ripeness in zone/terroir expression. Hunter, J.J., Pisciotta, A.,Voschenk, C.G., Archer, E., Novello, V., Deloire, A. & Nadal, M. Joint Conference (SASEV,OIV, GESCO) on Viticultural Zoning, 15-19 November 2004, Cape Town, South Africa.

Phenolic extraction during fermentation as affected by ripeness level of Syrah/R99 grapes.Nadal, M., Volschenk, C.G. & Hunter, JJ. Joint Conference (SASEV, OIV, GESCO) onViticultural Zoning, 15 -19 November 2004, Cape Town, South Africa.

Phototropic and geotropic shoot orientation: Effect on physiological, vegetative andreproductive parameters. Pisciotta, A., Di Lorenzo, R., Barbagallo, M.G., Volschenk, C.G. &Hunter, JJ . Joint Conference (SASEV, OIV, GESCO) on Viticultural Zoning, 1 5 - 1 9November 2004, Cape Town, South Africa.

Shoot positioning: Effect on physiological, vegetative and reproductive parameters. Pisciotta,A., Di Lorenzo, R., Barbagallo, M.G., Volschenk, C.G. & Hunter, J.J. Joint Conference(SASEV, OIV, GESCO) on Viticultural Zoning, 1 5 - 1 9 November 2004, Cape Town, SouthAfrica.

Sugar loading and phenolic accumulation as affected by ripeness level of Syrah/R99 grapes.Deloire, A. & Hunter, J.J. Joint Conference (SASEV, OIV, GESCO) on Viticultural Zoning, 15 -19 November 2004, Cape Town, South Africa.

Aspects concerning optimal grape ripeness and wine quality. Hunter, J.J., Volschenk, C.G.,Pisciotta, A., Archer, E., Nadal, M., Novello, V. & Deloire, A. VinPro Meeting. 27 June 2004,Stellenbosch.

Aspects concerning the determination of optimal ripeness. Hunter, J.J., Volschenk, C.G.,Pisciotta, A., Archer, E., Nadal, M., Novello, V. & Deloire, A. Seminar, University of Palermo,Marsala Campus, Sicily. May 2004.

Optimal ripeness and zone/terroir expression. Hunter, J.J. Seminar, University of Mendoza,Mendoza, Argentina. May 2005.

Influencia de la maduracion y calidad del vino tinto de Syrah. Varvaro, G., Nadal, M., Lopez,N. & Hunter, J.J. Gienol Symposium, 3 - 4 June 2005, Palencia, Spain.

Final report 27

4. Total cost summary of project

Total cost in real terms for year 1

Total cost in real terms for year 2

Total cost in real terms for year 3

Total cost in real terms for year 4

Total cost in real terms for year 5

TOTAL

Year

2002/2003

2003/2004

2004/2005

CFPA DFPT DFTS Winetech

143555

167305

184147

495007

THRIP TOTAL

175455

204484

225069

319010

371789

409216

1100015