• The Institute was awarded the prestigious McWilliam's Wines Maurice O'Shea Award in recognition of its contribution to the success of the Australian wine industry over the past 48 years.
  
• The Institute attracted one of the world's very best wine microbiologists, Professor Isak Pretorius, to the position of Director of Research with effect from January 2003. This augurs well for the future of the Institute and it is a reflection of the high esteem in which it is held. The strategic foresight of the GWRDC in assisting this critical development is acknowledged gratefully.
  
• Using organic synthesis we proved the structures of two malvidin-derived wine pigments and established their colour properties. With the help of these synthesised reference compounds we demonstrated that the pigments, which we had detected previously for the first time in red wine, were almost inert to SO₂-mediated bleaching, and that they showed superior colour properties and enhanced extinction coefficients in model wine at pH 3.6 when compared to grape-skin anthocyanins.
  
• A method for the measurement of acetaldehyde has been developed and validated. This enables us to monitor acetaldehyde concentrations in various tannin-, yeast- and wine oxidation-related research projects and will assist in developing a more detailed picture of this reactive aldehyde.
  
• With the help of rapid HPLC methods developed by the Tannin team, more than thousand samples from research projects of the Tannin, NIR and Microbiology teams have been analysed to establish anthocyanin, tannin and 'pigmented polymer' concentrations in samples from viticultural and winemaking trials. These data are vital to improve our understanding of the correlation between grape polyphenols, and the composition, colour and mouth-feel of wine. They also form the basis for our investigations into the potential of NIRS for the efficient measurement of polyphenols.
  
• Using solvent step-gradients with multilayer coil countercurrent chromatography (MLCCC), the isolation of anthocyanins and coumaroylated anthocyanins from commercially available grape marc extracts was optimised. In addition, MLCCC was instrumental in enriching very polar and potentially novel pigments for structure elucidation efforts, and in isolating 'pigmented polymers' for sensory studies.
  
• 2,3,6-Trimethylphenylbutadiene (TPB) has been identified as a hitherto unrecognised wine flavour compound that is formed by hydrolysis of grape-derived glycoconjugates. The threshold of TPB in a white wine was determined to be 40 ng/L and sensorial descriptors ranged from 'floral', 'geranium' and 'tobacco' at the lower concentrations to 'pungent', 'very green', 'unpleasant', 'plastic', and 'insecticide' at concentrations of 270 ng/L and above. This makes it one of the most potent wine flavour compounds known. Identification of TPB in wines has furthered our understanding of the contribution of grape secondary metabolites to wine quality.
  
• Four isomeric potential damascenone precursors, have been synthesised. The availability of these authentic samples allowed us to prove that they were intermediates in the formation of the important wine aroma compound damascenone. These findings have led us to re-evaluate the mechanism by which damascenone is formed in wine and other food products and will facilitate our goal of being able to improve wine quality by manipulation of wine composition in the vineyard and in the winery.
  
• Glucosides of the open-chain forms of cis- and trans-oak lactone have been synthesised. These compounds are stable in solution at wine pH, but are converted to the corresponding lactones at barrel-toasting temperatures. This demonstrates that such compounds can act as precursors of oak lactone during barrel manufacture, but not during subsequent wine conservation. By separating the isomeric glucosides from each other we have also been able to resolve the nature-identical isomer of cis-oak lactone from its non-natural optical isomer and evaluate its sensory properties. Previous sensory work on cis-oak lactone in wine has been carried out on mixtures of these two forms (i.e. the racemate), or even on mixtures of cis- and trans-oak lactone. The sensory detection threshold of the nature-identical isomer of cis-oak lactone (which is ten times more potent than the trans form) in a neutral dry white wine was 23 ug/L. This is much lower than the value of 90 ug/L previously reported for the racemic mixture.
  
• We have demonstrated that at wine pH, (3.0 - 3.8), oak lactones exist almost entirely in the aroma-active lactone form at equilibrium, in contrast to an earlier hypothesis. At the highest pHs studied, however, the open chain form of the cis-isomer lactonised relatively slowly, indicating the need for suitable soaking times and low pHs of wine or aqueous extracts when preparing extracts of oak shavings for analysis. The data show that wines treated with oak chips or shavings over a few days could continue to develop 'oakiness' by additional formation of cis-oak lactone once the chips or shavings are removed.
  
• Sensory studies robustly established that purified anthocyanin glucosides and their corresponding coumarates were not perceived as bitter or astringent. In addition, these grape pigments did not influence the fullness or viscosity of the solutions in which they were tasted.
  
• The effect of ascorbic acid addition on oxidation of wine continues to be a matter for debate in the literature. Previous laboratory scale work from the Institute (Institute publication no #577) and others indicated an enhanced depletion of sulfur dioxide in wine and model solutions to which ascorbic acid was added but formal sensory assessment has rarely been conducted in these experiments. Through collaboration with a winery we have now examined the effect of ascorbic acid addition to a Riesling and a Chardonnay wine bottled on a large scale under commercial conditions. Aroma, colour and compositional data indicate that for four closure types examined, these two wines were less oxidised after storage under ideal conditions for two and a half to three years when ascorbic acid was added at bottling. When analysed after six months storage there was little difference between the wines (Institute Annual Report 2000).
  
• The potential for mass spectrometry of red and white juice proteins to differentiate varieties of Vitis vinifera has been further confirmed. The extent to which the discriminating power can be extended to white wine is limited if bentonite fining has occurred. However, as DNA typing appears not to be practically possible for wine samples, the extension of our findings might represent one of very few avenues by which varietal authentication of white wine can be executed with some degree of certainty, if necessary through combination of analyses of secondary metabolites and, of course, sensory analyses for varietal typicity.
  
• Milligram quantities of a haze-protective factor (6xhis Hpf2p) have been overexpressed in yeast and purified. This has allowed us to treat a wine and assess its short to medium term stability during storage at 25^oC. After three and six months storage, 6xhis Hpf2p had retained full haze-protective activity, an important prerequisite for commercial viability in relation to the prevention of protein hazes in white wine.
  
• Methods to measure some volatile thiols of importance to varietal flavour have been developed and are now being applied to identify commercial wine strains that release either high or low amounts of these important aroma compounds. These findings will initially be important for modifying the flavour of a wide range of varietal wines.
  
• A collaboration with The Australian Proteomic Analysis Facility (APAF) in Sydney has permitted the identification of proteins that are differentially expressed in wine yeast strains. In the longer term, it is hoped that this research will identify factors that provide wine yeast with their important winemaking properties and is expected to lead to the generation of improved strains.
  
• Flavour metabolites that are produced at much higher concentrations in Saccharomyces species that are generally not associated with winemaking have been identified. These results indicate that interspecific hybrid yeast offer an effective strategy to harness diverse flavours in wine.
  
• Winery trials made in various wine regions with AWRI isolates of S. bayanus yeast are confirming our laboratory trials that have shown that these yeast can impart a greater diversity of sensory characteristics than achieved with S. cerevisiae, especially to aroma and palate fullness and texture. Demand for scaled-up winery trials with these S. bayanus yeast has increased to the point that we have engaged Lallemand to develop commercial starter cultures.
  
• A random amplified polymorphic DNA (RAPD) method for determining genetic similarity and differentiating strains of commercial and wine isolates of the malolactic bacterium Oenococcus oeni has shown that: (i) all commercial strains tested are genetically distinct; (ii) Australian isolates from different wineries are genetically distinct and at least as genetically diverse as the commercial strains; and (iii) isolates from a given winery may be genetically diverse.
  
• The adventitious growth of spoilage acetic acid bacteria in bottled red wine stored in an upright vertical, but not horizontal position, has led us to hypothesise that the upright storage position created a heterogeneous environment which allowed the growth of a bacterial film in only those bottles sealed with cork closures with substantial permeability to oxygen. Such a heterogeneous environment would likely not exist in horizontally stored bottles since the larger volume of wine adjacent to the cork would strongly compete with the bacteria for the oxygen as it diffuses through the cork closure. The direct measurement of oxygen content of spoiled and non-spoiled wines support this hypothesis which has been published in an international scientific journal (Institute publication #718), but still requires further rigorous confirmation.
  
• The Industry Services' team responded to 1,676 enquiries during the year, and analysed 2,231 samples (an 81% increase in sample analysis over the previous year).
  
• A record number of Roadshows were conducted, with ten seminars and five workshops being held in four states.
  
• The flow of new information from the Institute's closure trial continued, with three publications during the year (two in Technical Review and one conference paper [see Appendix 1]). Nineteen interviews were conducted with media and trial results were used or referred to in a number of other articles published both in Australia and overseas.
  
• The Institute's Dekkera/Brettanomyces control seminar was presented on 19 occasions during the year to approximately 530 winemakers, and has been presented to approximately 1550 winemakers since 1999. An ongoing targeted survey of the composition of Cabernet Sauvignon wines from five Australian wine regions indicates an apparent fall in the concentration of key Dekkera/Brettanomyces yeast derived spoilage compounds in wines from the 2001 vintage, compared to the previous five vintages. The basis and durability of this pleasing trend is to be investigated further.
  
• A trial conducted in conjunction with grapegrower and winemaker associations and wine companies into the effects of bushfire smoke on grapes and wines, elucidated an understanding of the nature of the problem and possible taint-minimisation strategies.
  
• Efforts to develop a practical NIR method for determining colour, total soluble solids (TSS) and pH in red grapes continue and have been outlined in an industry publication (Institute publication #730). During the 2003 vintage, the total number of samples analysed by NIRS represented approximately 300,000 tonnes of fruit, in excess of 20% of the total national harvest. Significant benefits have also been demonstrated from the use of NIR in the research environment. Using NIR to predict red grape quality parameters (colour, TSS and pH), researchers at CSIRO (Land & Water's Precision Agriculture group) and The University of Adelaide (Roseworthy campus) have been able to significantly reduce the cost of analysis and the necessity for wet chemistry in relation to their research projects.
  
• Preliminary results have indicated that neither freezing of red grape samples prior to homogenisation, nor the type of homogeniser had any significant effect on the accuracy of laboratory determinations of colour and TSS. However, as one might expect, there was an effect of freezing on pH. Work is continuing with further samples remaining in frozen storage for later analysis to investigate any long-term stability effects.
  
• The Institute has been involved in the testing and assisting in the development of an automated grape sample preparation system 'Bioprep5 Robotics', which has been released commercially by a local Adelaide company. The results from samples prepared with the automated system compared well to those obtained by the manual method in terms of accuracy and precision. The unit appears to offer potential in speeding up the preparation of red grape homogenates for scanning by NIR. (up to 112 samples per 8 hour day) when configured to act as an automated homogeniser.
  
• An evaluation of the feasibility of NIR spectroscopy for detection of botrytis and powdery mildew in grapes is continuing in collaboration with the Fungal Pathology Laboratory of The Adelaide University and CSIRO in Canberra. Distinct spectral changes were observed in fruit that had been visually classified into a number of different levels of powdery mildew infection. Principal component analysis of the spectra was used to detect and correctly classify, powdery mildew infection down to at least a 10% infection level. Similar work is also in progress with botrytis-infected fruit. The implication of this work is that it may be possible to discriminate fungal infected fruit at the weighbridge to provide a 'go/no-go' test to highlight suspect fruit for further detailed analysis to determine suitability for winemaking.
  
• Stable isotope dilution analysis methods for 33 important fermentation esters, acids and alcohols using their deuterium labelled analogues as internal standards have been fully validated. The methods have facilitated accurate and rapid measurement of the composition of wines and model media ferments and are are now being applied to a range of cross-disciplinary projects at the Institute including that of flavour scalping by closures.
  
• Stable isotope dilution analysis methods for wine taint compounds derived from petroleum products, including toluene, styrene, C2 alkyl benzenes (e.g. xylenes), C3 alkyl benzenes (e.g. trimethylbenzenes), C4 alkyl benzenes (e.g. tetramethylbenzenes), naphthalene, methylnaphthalenes, dimethylnaphthalenes and trimethylnaphthalenes have been fully validated. The methods enable us to more readily assess wines for possible contamination.
  
• The Viticulturist responded to 415 enquiries, participated in nine Institute Roadshows and five Research to Practice^TM workshops across three states.
  
• The Analytical Service laboratory underwent a technical assessment by NATA to the new ISO/IEC 17025 standard. Two new test methods; namely an enzymatic acetic acid assay and the HPLC multiresidue assay were accredited and two additional staff members have become NATA signatories.
  
• A new Analytical Service fee schedule has been printed as result of a review of the service delivery charges. In addition, an extensive market research report was commissioned to establish how we were meeting our client expectations. This report has been extremely positive, with no identifiable areas of concern. The results are consistent with increases in revenue in the past financial year. A large proportion of the Analytical Service activities are now associated with customised non-routine projects for both Australian and overseas companies.
  
• Eleven thousand copies of the Institute's annual publication, Agrochemicals registered for use in Australian Viticulture 2003/2004 were produced in response to an increase in demand. The reduction in the number of agrochemical related enquiries in comparison to previous years illustrates the effectiveness of the Institute's role in compiling and distributing this publication.
  
• A booklet entitled The A-Z of information on wine and health was developed and published. The Commonwealth Department of Health and Aged Care favourably reviewed the booklet and contributed towards its publication and printing costs. As of 30 June 2003, approximately 48,500 booklets have been distributed.
  
• Staff of the John Fornachon Memorial Library responded to 3,608 requests for information during 2002/2003.
  
• Over 5,000 new records were added to the web-accessible database of the Library (available only to Australian winemakers and grapegrowers) during the year, making a total of over 25,000 records available for searching, 24 hours per day, 7 days per week.
  
• Six issues of Technical Review published during the 2002/2003 financial year were consolidated onto a searchable CD ROM and distributed to Australian winemakers and grapegrowers. The new website of The Australian Wine Research Institute was launched to industry. The new website is easier to navigate and faster to download and contains additional useful information.
  

General highlights

• The Institute published 41 papers on Institute activities in refereed and non-refereed publications.
  
• Institute staff gave 174 oral presentations, conducted thirteen Workshops and presented 20 posters.
  
• Institute staff presented 48 lectures and coordinated a six week subject to undergraduate students.
  
• Institute staff supervised/co-supervised 31 postgraduate students.
  
• Institute staff recorded and responded to 6,001 requests for information during the 2002/2003 year or, to put the statistics into perspective, 24 people contacted the Institute seeking information on every working day of the year. This figure does not include request for work through the Analytical Service which conducted 54,000 individual analyses during 2003/2004.
  
  

Readers are strongly encouraged to read the Institute's 2003 Annual Report in detail rather than relying on the dot points above for information.