Source: UNIVERSITY OF CALIFORNIA, DAVIS submitted to
EVALUATION OF FACTORS AFFECTING AROMA AND TASTE PERCEPTION BY SENSORY AND INSTRUMENTAL METHODS
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
TERMINATED
Funding Source
Reporting Frequency
Annual
Accession No.
0066260
Grant No.
(N/A)
Project No.
CA-D*-VIT-3269-H
Proposal No.
(N/A)
Multistate No.
(N/A)
Program Code
(N/A)
Project Start Date
Jan 10, 1999
Project End Date
Sep 30, 2004
Grant Year
(N/A)
Project Director
Noble, A. C.
Recipient Organization
UNIVERSITY OF CALIFORNIA, DAVIS
410 MRAK HALL
DAVIS,CA 95616-8671
Performing Department
VITICULTURE AND ENOLOGY
Non Technical Summary
(N/A)
Animal Health Component
50%
Research Effort Categories
Basic
50%
Applied
50%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
50111311020100%
Knowledge Area
501 - New and Improved Food Processing Technologies;

Subject Of Investigation
1131 - Wine grapes;

Field Of Science
1020 - Physiology;
Goals / Objectives
1. Investigation of basic factors affecting sensory perception with ;emphasis on ;temporal perception of astringency, viscosity and taste stimuli. 2. Understanding the relationship between sensory and instrumental analysis of wine volatiles by collection of instrumental and sensory data and comparison of multivariate statistical methods for simplifying these data sets to find underlying dimensions and relate instrumental to sensory data. 3. Investigation of the use of alternative statistical methods to evaluate time-intensity and escriptive anlaysis data including distribution ;analyses, multivariate and meta-analytical statistical procedures.
Project Methods
Time-intensity evaluation of model solutions and wines will be used to evaluate factors affecting temporal perception of stimuli, such as judge training, difference in salivary flow and compound structure. Descriptive analyses of wine flavor will be related to instrumentalvolatile data for the same wines by several multivariate statistical techniques including Procrustes Analysis and Partial least Squares Regression to seek relationships and underlaying dimensions in the data.

Progress 01/10/99 to 09/30/04

Outputs
Using temporal methods, the effect of sweetness (increased by Aspartame [APM]) and of flavors (berry or "vegetative") on perception of astringency was evaluated in red wine. Astringency intensity of the sweetened wine was significantly lower than the unsweetened, while the added essences had no effect. The intensity of fruitiness perceived orally was increased significantly in red wine by sweetness of added APM; in contrast fruitiness intensity was not affected by increased intensity of berry flavor (addition of berry essence) or astringency (added grape seed tannin). In both cases, the decrease of astringency and increase in berry flavor produced by added sweetness was a cognitive effect since the addition of APM did not change viscosity nor the composition of the volatiles. Descriptive analysis of 28 sparkling wines after 7 years lees aging (SW-7) was compared with profiles of the base wines used in the cuvee (cuvee) and profiles of same sparkling wines after 2 years lees aging (SW-2). Caramel intensity increased in SW-2 made from white varieties (Chardonnay and Pinot blanc) over that of their cuvees, while berry aroma decreased in SW-2 made from red varieties (Pinot noir and Pinot Meunier). The intensity of all attributes in all 4 varietals increased from the cuvee to SW-7 wines. The difference in aroma between the SW-2 and SW-7 was also an increase in aroma of all terms. Overall, no consistent change in aromas was observed over time as a function of grape origin or variety from the cuvee to SW-7 or SW-2 to SW-7. Thus it is not possible to predict aroma of SW with extensive lees aging by variety nor from aroma profiles of their cuvees or of their SW with shorter lees aging times.

Impacts
Previously it has been shown that astringency can be decreased by the addition of sucrose which increases sweetness and viscosity. In other studies addition of polymers which increase viscosity have also decreased astringency. This study shows that astringency can be decreased by the cognitive effect of sweetness. Although 28 wines of 4 different varietals were studied, no consistent changes in aroma from the base cuvees to the aromas of the SW after 7 years of lees age were found. More importantly, aromas of SW made from different varietals could not be clustered or discriminated.

Publications

  • 2001 Noble, A.C. Sensory methods of flavour analysis. In Food Flavour Technology, ed: A.J.Taylor, Sheffield Academic Press, Sheffield, UK. p. 252-275.
  • 2002 Noble, A.C. and Ebeler, S.E. and Use of multivariate statistics in understanding wine flavour. Food Reviews International. 18,1-21.
  • 2002 Noble, A.C. Astringency and Bitterness of Flavonoid Phenols. In Chemistry of Taste: mechanisms, Behaviors, and Mimics, ACS Symposium Series # 825, Amer. Chem. Soc. Washington, D.C. (Ed. Given, P. and Paredes, D.) pp.192-201.
  • 2002 Naim, M., A.I. Spielman, A.C. Noble, I. Peri, S. Rodin, and M. Samuelov-Zubare. The hypothesis of receptor- dependent and receptor independent mechanisms for bitter and sweet taste transfuction: Implications for slow taste onset and lingering aftertaste. In Chemistry of Taste: mechanisms, Behaviors, and Mimics, ACS Symposium Series # 825, Amer. Chem. Soc. Washington, D.C. (Ed. Given, P. and Paredes, D) 2-17.
  • 2002 Noble, A.C. Wine Tasting. Encyclopedia of Food Sciences and Nutrition. Ed. B. Caballero, L, Trugo and P. Finglas. Elsevier. London p.4942-4946.
  • 2002 Froest, M.B. and Noble, A.C. Effect of wine flavor and level of consumer wine knowledge and sensory acuity on preference for young red wines American Journal of Enology and Viticulture 53:275-284


Progress 01/01/01 to 12/31/01

Outputs
Astringency intensity increases upon successive sips of the same wine or astringent stimuli and of stimuli sampled within a session. To determine which rinse best reduces this carryover effect, astringency (A) of red wine was rated continuously while wine was sipped, spit (at 10s) and rinses(or crackers) were sipped (or chewed) at 20s and spit at 30 s. Pectin (at1 or 5g/L) reduced A after 30s more than CMC (0.01 or 1.0 g/L), Polycose (5 or 40g/L) or water. Pectin (5g/L) reduced A after 30 s more than crackers (unsalted oyster) or water. Although water lowered A from 30 to 40s more than crackers, after 40 s crackers were more effective. Consistent with previous studies, astringency was rated less intense when artificial saliva containing salts (ArtSalt) was introduced at high flow rates (5 or 8ml/min) than at a low flow rate (0.5ml/min). When stimuli were sipped repeatedly, the maximum intensity (IMAX) of A increased with each sip at low flow rate, but IMAX was lower and decreased upon successive sips at high flow rates. When 3g/L gelatin was added to the ArtSalt (ArtSaltProt), the same effect was observed. However, there was no difference in astringency IMAX between ArtSalt and ArtSaltProt at either flow rate for any sip. 126 consumers rated 10 identified wines (previously profiled by descriptive analysis) for purchase intent (PI) and also rated the influence of concepts (price, brand, label description etc) on PI. Cluster analysis of PI revealed 4 consumer segments. Segments 1 and 2 were more influenced by brand, 3 by label description and 4 by label and price. Segment 1 and 2 rated PI highest for wines whose brands were also rated most influential in the concept test. Segment 3 rated a wine with fruity aroma highest in PI while a label description of "fruity flavor" was highest in the concept test. Segment 4 similarly rated PI highest for a moderately oaky and fruity wine, while a label describing oak aging was most influential in the concept test.

Impacts
Without reduction in carryover of A, many wines cannot be tasted in one session with valid results. The use of pectin between samples alleviates this problem considerably. The demonstration that crackers have a significant effect in reducing A, validates their common usuage during tastings, although pectin is far more effective. The correspondence of results from consumers rating of purchase intent with the results obtained in concept tests suggests that this rapid non-fatiguing method could be used for evaluating factors influencing consumer preference (and wine sales).

Publications

  • Cubero-Castillo, E., Noble, A.C. (2001). Effect of compound sequence on bitterness enhancement Chem Senses. 26:419-424.
  • Nygren, I.T., Gustafsson, I.B., Haglund, A. Johansson, L., and Noble A.C. (2000) Flavor changes produced by wine and food interactions. Chardonnay wine and Hollandaise Sauce. J. Sensory Studies 16: 451-470.
  • Noble, A.C. Sensory Evaluation in the wine industry, an under-utilized resource. Proceedings ASEV 50th Anniversary Annual meeting. Seattle, WA. June 19-23, 2000. pp 1-2. Am. J. Enol. Vitic. Davis, CA.
  • Yegge, J. M. and Noble, A C Identification of sensory and non-sensory attributes of Californian Chardonnay wines that influence acceptance and purchase intent for differing segments of consumers. Proceedings ASEV 50th Anniversary Annual meeting. Seattle, WA. June 19-23, 2000. pp 28-31. Am. J. Enol. Vitic. Davis, CA.
  • Ebeler, S.E. and Noble, A C Past and the future: bucket flavor chemistry to Senso-chemistry. Proceedings ASEV 50th Anniversary Annual meeting. Seattle, WA. June 19-23, 2000. pp 205-208. Am. J. Enol. Vitic. Davis, CA.
  • Brossaud, F., V. Cheynier, and A C Noble. (2000) Bitterness and astringency of grape and wine polyphenols Austr. J. Grape and Wine Research 7, 33-39.


Progress 01/01/00 to 12/31/00

Outputs
Astringency of red wines was rated continuously while wines were sipped at 25 or 30 s intervals in 2 different studies. The maximum intensity of astringency with 25 s increased 20 % with each successive sip, where as it only increased 5 % for 30 s. The perceived astringency of several wines sipped at 25 and 35 s is being rated to compare reduction in carryover. Several different rinses are being tested to find the one which minimizes this best. In previous studies, astringency was rated less intense by subjects (Ss) with high salivary flow rates. This was confirmed in a study removing variability in scale use between Ss. When artificial saliva was introduced, astringency of tannic acid and alum solutions decreased as the flow rate was increased. 120 consumers rated preference for flavor of 10 Chardonnay wines. Five clusters with difference preferences were found, of which 3 could be modeled. One group preferred wines more as floral fruity notes increased, with the other as the vanilla, spicy notes increased. The third group preferred wines with specific balance of the flavors. For consumers in this study, and 600 questionnaire respondents, the most important non-sensory factors influencing wine purchase were flavor and the variety, with the label and bottle least so. Despite this, preference for one wine significantly changed when the duplicates were identified as from two different wineries.

Impacts
Without development of a valid procedure, the carryover effect renders evaluations of red wines useless. An improved protocol will be valuable for tasting by winemakers informally or by wine judges at a show. Non-sensory and sensory factors influence consumer acceptance and purchase intent. Understanding these will permit wineries to increase market share.

Publications

  • Pfeiffer, J.F., Boulton, R.B. and Noble, A.C. 2000. Modeling the sweetness response using time Intensity Data. Food Qual. Pref. 11:129-138
  • Park, S.K., Boulton, R.B. and Noble, A.C. 2000. Automated HPLC analysis of glutathione and thiol-containing compounds in grape juice and wine using pre-column derivatization with fluorescence detection. Food Chem. 68(4):475-480.
  • Park, S.K., Boulton, R.B. and Noble, A.C. 2000. Formation of hydrogen sulfide and glutathione during fermentation of white grape musts. Am. J. Enol. Vit.51(2):91-97.
  • Dewey, F.M, Adams, D.O, Ebeler, S.E., Noble, A.C. and Meyer, U. 2000. Quantification of Botrytis in grape crush determined by a monoclonal antibody-based immunoassay. Am. J. Enol Vit. 51(3):272-282.
  • Noble, A.C. 2000. Wine Tasting. Encyclopedia of Food Science and Nutrition. Elsevier. London


Progress 01/01/99 to 12/31/99

Outputs
Thresholds of propyl thiouracil (PROP) were correlated with those of sucrose octaacetate (SOA), naringin (N), and caffeine (C) but only for women. Thresholds for six bitterants and for PROP were uncorrelated to intensity ratings of the six compounds. Compound specific carry-over effects were quantified pairwise showing that caffeine in the first position increases the bitterness response of SOA, N, C, denatonium (D), quinine(Q) and limonin(L), while none increased bitterness (Bit) of C. Bit of Q and D was increased significantly when any of six compounds preceded them. Astringency (Ast) of tannic acid, alum and black tea was rated continuously throughout eight sips. Ast increased with each sip, although the increase was no longer significant after the fourth to fifth sip. Sourness, Bit and Ast were rated by time intensity methods in twelve red wines. Winemakers (WM) and students grouped the wines by similarity of mouthfeel, and rated 'hardness' (hard). Maximum intensity (IMAX) of Ast was correlated significantly with PPI, Bit IMAX, phenols, and Hard. The wines were not grouped by Ast, Bit, Sour, Hard by either WM or students

Impacts
Individuals perceive bitterness of bitter compounds such as caffeine (coffee/tea), quinine (tonic), limonin and naringin (citrus) very differently. Test of treatments of additives to mask bitterness in foods and beverages must examine the responses of individual subjects before averaging the data. For both bitterness and astringency, when the same wine or sample is sipped twice, the second sip is more bitter or astringent due to 'carry-over' effect. Unless thorough rinsing is used, comparison of two bitter or astringent samples cannot be done using pair tests.

Publications

  • Peleg, H., Gacon, K. and Noble, A.C. 1999. Bitterness and Astringency of Flavan-3-ol Monomers, Dimers and Trimers. J. Sci. Food Agric. 79:1123-1128.
  • Courregelongue, S., Schlich, P. and Noble, A.C. 1999. Using repeated ingestion to determine the effect of sweetness, viscosity and oiliness on temporal perception of soymilk astringency. Food Qual. Pref. 10(4-5):273-279
  • Noble, A.C. and Pfeiffer, J.F. 1999. Wine. In Francis, F.J. (ed). The Wiley Encyclopedia of Food Science and Technology, 2nd Ed. Wiley, NY,
  • Peleg, H. and Noble, A.C. 1999. Effect of composition, temperature and pH on astringency in fruit juice. Food Qual. Pref. 10(4-5):343-347.
  • Noble, A.C. 1999. Using analytical sensory techniques to understand wine preference. In Bell, G. and Watson, A (eds.). Tastes and Aromas. The Chemical Senses in Science and Industry. A.University of New South Wales Press, Sydney. pp. 98-104.


Progress 01/01/98 to 12/31/98

Outputs
The volume of a typical sip of wine (10ml) and time between sips (18-32 s) were determined. Using a repeated sip time intensity (TI) protocol (4 sips of10 ml wine at 25 sec intervals which were swallowed 6 s after each sip), maximum intensity of astringency (Imax) increased upon successive sips as did minimum intensity between sips. For 4 red wines, the rate of increase from sip1 to 4 correlated with Imax of sip 1. Using a 30 s sip interval, astringency Imax of a fruit juice beverage (pH 2.7) decreased slightly from sip 1 to sip 3. Using single sip T-I, the astringency and bitterness of Cabernet Franc grape seed tannin was equal to that of grape skin tannin (both 1500mg/L) in either water or wine base. Addition of anthocyanin to the tannin fractions had no affect on either attribute. Thresholds of 41 subjects for propyl thiouracil (PROP) were correlated with those of sucrose octa-acetate (SOA), naringin, and caffeine, but not with limonin, quinine, and denatonium benzoate, showing that sensitivity to PROP cannot be generalized to all bitterants. IMax (from TI of suprathreshold levels of these compounds) were correlated with their respective thresholds for caffeine and PROP only, whereas duration of aftertaste was correlated with threshold only for PROP and quinine. Descriptive analysis (DA) of the effect of food on wine and vice versa showed that the wines had little effect on the food flavor, but conversely, the wine flavor was altered after the food.

Impacts
(N/A)

Publications

  • SMITH, A. K. and NOBLE, A. C. 1998. Effects of increased viscosity on the sourness and astringency of aluminum sulfate and citric acid. Food Quality and Preference 9(3):139-144.
  • PELEG, H., BODINE, K. and NOBLE, A. C. 1998. The influence of acid on astringency of alum and phenolic compounds. Chem Senses
  • SWOALSKY, R. A. and NOBLE, A. C. 1998. Comparison of effect of the concentration, pH and anion species on the sourness and astringency of organic acids. Chem Senses 23(3):343-350.
  • de la PRESA OWEN, C., SCHLICH, P., DAVIES, H. D. and NOBLE, A. C. 1998. Effect of secondary fermentation of aroma of four V. vinifera varieties. Am J Enology and Viticulture 49(3):289-294.


Progress 01/01/97 to 12/01/97

Outputs
Descriptive analysis (DA) of base wines, Pinot blanc and P. Chardonnay wines were separated from the red varietals, P. noir and P. Meunier. In contrast, DA of the sparkling wines after 18 months on lees showed no clustering by variety or grape color. Principal component analysis (PCA) of the difference in scores between each base and its sparkling wine, showed clustering by grape color due to larger increase in citrus and apple intensity in red varietals and of vanilla/butter aroma in whites. Astringency of aqueous solutions of phenolic compounds (grape seed tannin, tannic acid, catechin and gallic acid) increased upon addition of citric acid, whereas the astringency of alum was reduced. Astringency of alum was decreased equivalently by addition of equi-sour levels of lactic acid, citric acid or hydrochloric acid. Sequential sipping time intensity (TI) methodology was used to measure the carry over effect of astringency upon repeated ingestion of soy milk. Maximum astringency increased upon successive sips for each treatment; raising viscosity with CMC reduced astringency more than sucrose addition, while oil had no effect. The rate of carbon dioxide (CO2) evolution from sparkling wine was monitored gravimetrically. Although CO2 evolved 2.5 times more rapidly from etched than unetched glasses, no significant difference in aroma was found between wines in etched vs unetched glasses.

Impacts
(N/A)

Publications

  • DE LA PRESA OWENS, C. P. and NOBLE, A.C. 1997. Effect of heated storage on aroma of oaked and unoaked Chardonnay wine. American Journal of Enology and Viticulture 48:310-316.
  • NOBLE, A. C. 1997. Missing the point. The Catch-22 of scoring wine quality. SLOW The International Herald of Tastes 6:14-17.
  • SMITH, A. K. and NOBLE, A. C. 1997. Effects of increased viscosity on the sourness and astringency of aluminum sulfate and citric acid. Food Quality and Preference 7:(in press).
  • DE LA PRESA OWENS, C. P., SCHLICH, P. , DAVIES, H. D. , and NOBLE, A. C. 1997. Effect of secondary fermentation of aroma of four V. vinifera varieties. American Journal of Enology and Viticulture


Progress 01/01/96 to 12/30/96

Outputs
The separate effects of titratable acidity (TA) and of pH on perceived astringency (AST) of organic acids were explored for citric, tartaric and lactic acids. Decreasing pH (while holding total acid constant) increased AST and sourness (SOUR); increasing total acid, while titrating to a constant pH, increased SOUR but had no affect on AST. Volatile compounds hydrolyzed (HOH) from skin (SK) and juice (Ju) extracts of Zinfandel grapes were compared by descriptive analysis (DA). Ju extracts were a slightly more potent source of aromas than SK. Little difference was observed between extracts HOH for one vs two months at 45C. Storage for 3 months increased tea, leather and butterscotch notes, while decreasing honey and prune raisin aromas. DA of commercially bottled Chardonnay wines held at elevated temperature showed the same changes in aroma (decrease in fruity notes and increase in tobacco and tea terms) as those reported previously for the same wine held in Schott bottles (at low oxygen), despite greater variation in oxygen levels. The wines were sorted on the basis of similarity (SIM) aroma before and after DA. No difference was found between the SIM space (from multidimensional scaling MDS of the sorting data) before and after DA. The first dimensions of the sorting spaces were only correlated with the 4th Principal component of the DA data-which only explained 5% of the variance. DA of base wines used for Sparkling Cuvees showed wines from white grape varietals (P. blanc and P.

Impacts
(N/A)

Publications

  • SMITH, A. K., JUNE, H. and A.C. NOBLE 1996. Effects of viscosity on the bitterness and astringency of grape seed tannin. Food Qual. Pref 7:161-166.
  • NOBLE, A. C. 1996. Introductory remarks to Special Issue of Food Qual Pref 7:149.
  • SMITH, A.K. and A.C. NOBLE. 1996. Effects of increased viscosity on the sournessan astringency of aluminum sulfate and citric acid. Food Qual Pref (in press).
  • NOBLE, A. C. 1996. Aroma-Taste Interactions. Trends Food Sci. Technol. 7:439-444.


Progress 01/01/95 to 12/30/95

Outputs
The separate effects of titratable acidity (TA) and of pH on perceived astringency of alum were explored using two organic acids. For both lactic and tartaric acids, increasing acidity by varying pH or TA decreased astringency maximum intensity (Max) and total duration (Dur), while increasing sourness Max and Tot. The sensory significance of volatile compounds hydrolyzed from glycosides extracted from skin (SK) and juice (Ju) or Merlot (M) and Cabernet Sauvignon (CS) grapes from California (CA) and Australia (Aus) were quantitatively compared by descriptive analysis (DA). SK extracts were a more potent source of aromas than Ju. Varietal and location differences were also observed. CS glycosides produced more intense fig, chocolate and tobacco aromas than M. Aus extracts were higher in tobacco notes, while CA were higher in berry and chocolate. Storage of wines at elevated temperature was monitored by sensory tests. Significant differences in aromas of wines held at 40C were produced in 7 to 9 days for white and 4 days for red. Both oaked and unoaked Chardonnays held for 30 and 45 days at 40C showed significant increases in intensity of honey and cooked (tea, tobacco and rubber) notes, and decrease in fruity (citrus, green apple and tropical fruit) and floral aromas. Storage for 15 days decreased both fruity and floral aroma intensities from those in the control wines, but did not produce an increase in the cooked notes.

Impacts
(N/A)

Publications


    Progress 01/01/94 to 12/30/94

    Outputs
    The effects of viscosity on perception of taste and mouthfeel were evaluated in several systems using temporal methods. In time-intensity evaluation (TI) of solutions varying in sweetness (Sw) and viscosity, Sw perception was not altered as VI was raised from 1.0 to 60 cP with a tasteless gum. Conversely, increasing glucose from 0 to 140g/L did not affect perception of Vi. No difference in perception of either Sw or Vi was found between subjects with low (LO) and high (HI) salivary flow. Increasing sucrose in red wine from 0 to 200 g/L (which increased Vi by <3 cP) depressed astringency (Ast) intensity at maximum (MAX) and Ast duration (DUR); increasing grape seed tannin (GST) by 500 mg/L did not affect Sw MAX or DUR. HI judges had significantly shorter DUR and lower MAX Ast, but no difference in perception of Sw occurred as a function of salivary flow. Increasing Vi by 7 cP in aqueous solutions of GST significantly depressed Ast MAX and DUR, but did not alter bitterness. Further reduction in Ast was observed when Vi was raised to 16 cP, with no further change observed as VI was raised to 45 cP. Sourness and Ast of citric acid (Ci) and alum (Al) were decreased by Vi increases > 5 cP. Ast of of equi-sour Ci, lactic and hydrochloric acid (HCl) solutions was depressed when each acid was combined with Al. At equisour levels, HCl was the most astringent acid. Ast of Al was also depressed when combined with each acid, suggesting that Al elicits Ast by a different mechanism than phenolic compounds.

    Impacts
    (N/A)

    Publications


      Progress 01/01/93 to 12/30/93

      Outputs
      Residual glucose (RG) levels in saliva (S) were determined at 30 sec intervals after administration of glucose solutions. RG concentration did not correlate with the total duration of sweetness (Sw) recorded in temporal evaluation (TI) of the same stimuli by the same subjects. Increasing viscosity (V) of gums had no effect on eliciting S flow, whereas increasing levels of sugar solutions of the same V increased Sw and S flow. TI of hydroxyl (OH) benzoic acid isomers which varied in the location and number of OH differed in Sw, bitterness, astringency, sourness and prickling. By descriptive analysis, Spanish wine varietals were profiled and discriminated. Macabeo was higher in nutty and "shoe polish" aromas. Parellada was highest in citrus, while Xarel.lo was highest in black pepper and tropical fruit. Analysis in Australia by GC-MS, using isotopically labeled standards, revealed that levels of 2-methoxy-3-isobutylpyrazine (MIP) in Cabernet Sauvignon wines were directly correlated with intensity of vegetative aroma and inversely to available light (Lite) within the vine canopy. Wine from the most vigorous site with least Lite contained 40 ppb MIP, in contrast to 4 ppb for vines with open canopies. Using a method for terpene analysis, high levels of phenylethanol (PEA) glycoside were found in wine suggesting possible glycoslyation of PEA by yeast.

      Impacts
      (N/A)

      Publications


        Progress 01/01/92 to 12/30/92

        Outputs
        The use of repeated measures (RM) analysis of variance (AOV) in examination of sensory data to compare the effect of a grouping variable was explored. Using RM AOV, it was demonstrated that no significant difference in temporal perception of sourness, bitterness or astringency, nor of sweetness, fruitiness or sourness occurred as a function of difference in salivary flow rate or ability to taste PROP. In contrast, small differences among salivary flow rate groups were found using a fixed model AOV. Using time-intensity methods, equimolar concentrations of epicatechin had a higher maximum intensity and longer duration of bitterness than its isomer, catechin, with only small differences in astringency. Descriptive analysis of Cabernet Sauvignon (CS) wines from vines on older soils with higher levels of clay, showed these wines to be higher in intensity of vegetative aroma and flavor by mouth, than vines grown in the same microclimate on other soils, all with lower water holding capacity. Pinot noir seeds had similar levels of polymeric flavor-3-ols and patterns of galloyl esterification as CS. However, P. noir skins lacked an anthocyanin-tannin adduct, present in other varieties. Sulfur containing compounds were monitored in fermentations (F) with and without added Glutathione (GSH). In contrast to model systems (M), addition of GSH increased production of hydrogen sulfide during F. Similar to M, GSH increased F rates in wine.

        Impacts
        (N/A)

        Publications


          Progress 01/01/91 to 12/30/91

          Outputs
          Bitterness (B), sourness (So) and astringency (A) were rated by time-intensity methods (TI) in water and in wine. No significant differences (Diff) were found between expectoration and swallowing except for A in wine. When samples were swallowed, A was lower and the total duration (DUR) was shorter. When TI responses for subjects with low (L), medium (M) and high (H) salivary flow rates were compared, inconsistent Diff were seen across groups. Where Diff were observed, in contrast to a previous study, H tended to have longer DUR of B or A than L, although L rated the samples higher in both attributes. No Diff were seen for So. Sweetness (Sw), fruitiness (F), and So were evaluated by TI in beverages sweetened with aspartame (APM) or sucrose (SUC). For both, fruitiness intensity increased as sweetner or acid levels were raised. F lasted longer in APM samples than SUC, although samples were equisweet. No differences in Sw, So or F were observed as a function of salivary flow rate. The lowest sweetner level elicited a greater salivary flow rate than the higher sweetness levels. Sulfur containing volatiles and non-volatile thiols were monitored in fermentig grape juice (GJ) and model systems (Model) by Gas Chromatography (GC) and HPLC, respectively. Glutathione (GSH) added to Model increased the rate of fermentation, and depressed the production of hydrogen sulfide (H2S). No differences among 4 yeast strains in production of H2S were seen in fermentations of GJ, despite differences in rates of fermentation.

          Impacts
          (N/A)

          Publications


            Progress 01/01/90 to 12/30/90

            Outputs
            Bitterness (B) and astringency (A) were rated by time-intensity methods (T-I) insystems varying in ethanol, pH or phenol (Tannic acid ?TA? or Catechin ?C?), Maximum intensity (MAX), total duration (DUR) of B increased with ethanol concentration, whereas A was unaffected; MAX and DUR for A and B increased more for TA than C; lowering pH had no affect on B, but slightly increased A. Pooled overall samples, subjects (S) with higher salivary flow had shorter time to MAX and DUR for both B and A. In a T-I study of equi-sweet sucrose and aspartame orange-flavored solutions, whether samples were unacidified, acidified (H+) or H+ and carbonated, fruitiness (F) and sweetness (Sw) had longer DUR in the aspartame solutions. By T-I, fruit drinks which were thickened with pectin, showed an increase in MAX and DUR of both Sw and F as viscosity increased. In comparing the effect of expectoration vs swallowing by T-I, although inconsistent differences were observed across different sweeteners, overall Sw persisted two seconds longer when samples were spit. S and F DUR was longer by Spit for sucrose sweetened drinks, whereas no difference in S or F DUR was seen for Aspartame. By headspace Gas Chromatography, in wines with "off-odors" and in model fermentations, the most frequently occurring sulfur-containing volatiles were hydrogen sulfide, dimethyl sulfide, diethyl disulfide, methanethiol and ethanethiol. No correlation among the compounds was observed.

            Impacts
            (N/A)

            Publications


              Progress 01/01/89 to 12/30/89

              Outputs
              Free (FT) and glycosidically bound (BT) terpenes in the skin and mesocarp of Symphony and Muscat (M) grapes were examined in the 1988 season. Eighty to 90% of the terpenes were BT with the remainder FT; FT and BT, which increased in concentration as the grapes developed, were evenly distributed between the skin and pulp. On fermentation of M grapes, FT increased from 118 to 750 mg/L. Descritive analysis of 3 vintages of commercial Cabernet Sauvignon wines was related to descriptive data for the soils from which the grapes originated by partial least squares regression analysis. No clustering within the 6 regions studied was observed, however, vegetative flavors were associated with soils high in clay, with a high water holding capacity (WHC); berry (fruity) aromas were associated with soils with high % of gravel, low WHC and low amount of soil in the root zone. Bittererness (B) and Sourness (S) intensity were rated wines varying in ethanol, pH, and catechin. Raising ethanol by 3% produced a larger increase in B than addition of 1400 mg of catechin. The pH increase from 2.9 to 3.2 also enhanced B, though a further rise diminished it. S decreased with increasing pH, but was unaffected by addition of ethanol or catechin. In a 2nd study, time to maximum intensity (MAX), MAX and duration (DUR) of B increased with ethanol concentration, while MAX and DUR for B and Astingency (A) increased upon addition of catechin and tannic acid. A, which persisted far longer than B, was slightly decreased by addition of ethanol.

              Impacts
              (N/A)

              Publications


                Progress 01/01/88 to 12/30/88

                Outputs
                Wines bottled in clear and green glass developed sunstruck (SUN) aromas after exposure to fluorescent light for 3.4.and 18 hrs (still wines) and 3.3 and 31 hrs (sparkling wines). SUN aroma was characterized by decrease in citrus and increase in cooked cabbage, soy and wet wool intensities. Time-intensity curves are being examined by Partial Least Squares Regression (PLS) to remove individual variation to see underlying dimensions in the temporal response curves. Canonical variates analysis (CVA) was compared to Principal Components analysis (PCA) using two sets of wine descriptive analysis (DA) data. CVA, using wines as the classification variable gave results very similar to PCA. However, classifying the wines by region or vintage, although separating the classes, altered the configurations. To evaluate the contribution of geography and soil to wine aroma, selected sites in 6 regions in Napa Valley are being studied. Trace element analysis of the soils and berries are being compared, with the soils being analyzed by standard descriptive and chemical methods. Sensory DA of 1986 wines reveals no regional clustering. DA of wines from 1987 and 1988 is in progress. Free and glycosidically bound terpenes in the mesocarp and epidermis of Symphony (S) and Muscat of Alexandria (M) grapes were examined over the 1987 and 1988 seasons. In 1987, most of the terpenes showed an increase in concentration with time. For all but nerol, higher amounts were found in the mesocarp. S grapes had less than 1/3 of the terpenes found in M.

                Impacts
                (N/A)

                Publications


                  Progress 01/01/87 to 12/30/87

                  Outputs
                  Grape musts of Gewurztraminer, White Riesling, and Muscat of ALexandria are being analyzed at several different maturity levels to follow the accumulation of terpenes and terpenic glycosides. Wines made from the same musts are also being analyzed to evaluate the effect of fermentation on these compounds. A computerized time-intensity (TI) system was developed and used to evaluate the intensity over time of sweetness (S) and fruitiness (F) in solutions varying in concentrations of glucose and peach extract and presented at two solution temperatures. For both attributes, increases in concentration of glucose and peach extract, respectively, increased the total duration, maximum (Max) intensity (I), time to Max, area under the curve (Area) and rate of onset. A further increase in Max I, total duration and Area for S was observed as the concentration of peach extract was raised; in contrast an increase in glucose concentration did not increase Max I of F, although increases in total duration and Area were produced. For S as the temperature of solutions was raised from 5 to 25 C, a significant increase in Max. I and decrease in time to Max. and lag time elicited. In contrast, temperature had no effect on the TI parameters for F.

                  Impacts
                  (N/A)

                  Publications


                    Progress 01/01/86 to 12/30/86

                    Outputs
                    Aroma thresholds (ug/1) were determined for 5 sulfur volatiles in white wine: dimethyl sulfide (DMS) 25, dimethyl disulfide 29, diethyl sulfide 0.92, diethyl disulfide 4.3, and ethanethiol (ETSH) 1.1. By descriptive analysis, the aromas of wines spiked with DMS or ETSH were profiled. DMS increased the intensity of asparagus, canned corn and molasses attributes, while ETSH increased the intensity of onion and rubber aromas. While on sabbatical leave, blycosidic precursors (GP) were isolated from Chardonnay (C) and Muscat of Alexandria (M) wines. In water, GP were characterized as having very low intensity fruity of floral flavors-by-mouth. GP were not bitter even at ten times the concentration present in wine. GP did not produce significant difference in wine taste even at 2 fold concentration (2X). Volatile aglycons, bound to the GP were liberated enzymatically. When added back to the wines, the M volatiles produced a significant difference in the aroma of M wine at 2X, while the C volatiles could not be detected until added to C wine at 3X.

                    Impacts
                    (N/A)

                    Publications


                      Progress 01/01/85 to 12/30/85

                      Outputs
                      For quantification of sulfur volatiles, 100 ul of wine headspace was injected directly on a gas chromatographic (GC) column, a section of which was chilled. Volatiles were detected by a flame photometric detector (FPO). This cryogenic procedure permits reproducible introduction and quantification of dimethyl sulfide, which was not possible by our previous procedure. For white wines aged in oak or red wines, volatile phenols recovered from wine by steam distillation co-elute with methoxypyrazines (MP) during HPLC. The steam distillate for these wines is being analyzed by GC using a nitrogen detector. Evidence for the photodegradation of MP has been obtained. After exposure to light for 120 hrs, aqueous solutions showed a loss of approximately 30%, whereas samples from which light was excluded showed no change. 163 commercial white wines were analyzed by descriptive evaluation. Intensity of five aroma attributes (floral, peach, citrus, green pepper, vanilla) and two taste terms (bitter and sweet) were rated by trained judges. Principal component analyses (PCA) of the mean intensity ratings showed considerable overlap among the varietal wines, although some discrimination was possible.

                      Impacts
                      (N/A)

                      Publications


                        Progress 01/01/84 to 12/30/84

                        Outputs
                        The temporal sequence of bitterness (B) and astringency (A) of four phenolic compounds were evaluated by time-intensity scaling (T-I) in white wine. As the duration of aftertaste(AT) increased, the maximum intensity(Max-I) increased, while the time to Max-I was generally independent of the intensity of either A or B. The rate of onset of A was faster for tannic acid, while the rates of decay for either A or B were the same for all compounds. Bitterness of caffeine(C) and quinine(Q) was evaluated in water by T-I. Max-I of B increased exponentially for C and linearly for Q. The duration of AT increased as the Max-I of B increased. For equi-bitter solutions, the AT persisted longer for C. C elicited a faster rate of onset and slower rate of decay of B. Two methods of analysis were developed. 1. Sulfur volatiles, trapped on Chromosorb 105, were analyzed by Gas Chromatography (GC) using a sulfur-specific Flame Photometric Detector(FPD). The range of linear response for dimethyl disulfide, diethyl sulfide and diethyl disulfide was 1 to 60 ng. Limits of detection were below the corresponding sensory thresholds. The coefficient of variation of recovery was less than 10% for several compounds including the 3 cited above. 2. Methoxypyrazines(P) were recovered from wines by steam-distillation. After clean up on C-18 Sep-Pak, P were quantified by reversed phase HPLC using a UV detector. Recovery is 60% or less for vines containing less than 1.5 ug/L.

                        Impacts
                        (N/A)

                        Publications


                          Progress 01/01/83 to 12/30/83

                          Outputs
                          Bitterness (B) and astringency (A) of four phenolic compounds were evaluated in white wine by two sensory methods: conventional intensity scaling and time-intensity scaling (TI). Using I, as the concentration of each compound was raised, the intensities of both B and A increased. Tannic acid and grape seed tannin were assigned higher ratings for A than B. In contrast, sallic acid and catechin solutions were rated more bitter than astringent, with B increasing more rapidly than A in both systems. Results from the TI study, in which the temporal aspects of B and A were measured, are being processed now. Using a vermouth flavored wine, sweetness (S), perceived viscosity (V) and bitterness (B) were rated at 4 sucrose (SU) levels (I-IV), in samples with the same 4 SU concentrations but with viscosity adjusted to that of IV, and in samples with SU concentration I, but with viscosity adjusted to that of the SU solutions I-IV. As viscosity increased, perceived S and V increased, while B decreased in the samples at constant SU concentration. S, V, and B ratings for the SU series I-IV versus those for the samples with the same SU concentrations, but with viscosities adjusted that that of IV, showed the same effect: within each pair with the same SU concentration, the more viscous sample was rated higher in S and V and lower in B. Standardized wine aroma terminology was proposed for the wine industry (Pub. #6).

                          Impacts
                          (N/A)

                          Publications


                            Progress 01/01/82 to 12/30/82

                            Outputs
                            Cabernet Sauvignon and Sauvignon blanc grapevines were examined for their physiological and enological responses to cluster thinning prior to and following veraison. Harvest yields of cluster thinned vines (TH) were 45 - 64% of the unthinned vines (Control). TH juices harvested on the same date were higher in pH, K and soluble solids. TH wines (made from grapes harvested on the same date) were higher in pH, K, total phenols, and for Cabernet Sauvignon color density and anthocyanins. Pre-and post veraison thinning produced comparable results in vine yields and must and wine composition. During maturation, K, titratable acidity (TiA), malic and tartaric acid, glucose, fruictose, free amino acid and proline were determined and expressed on both concentration and per berry bases. Results are still being analyzed. Perceived sourness intensity and parotid salivary flow in response to tartaric-fumaric acid solutions varying in pH(3.0-3.75) at a constant TiA(4.0g. tartaric/L) and in TiA(3.7 - 4.6 g./L) at a constant pH (3.5) were determined. Both sourness intensity (SI) and salivary flow (SAL) increased with TiA and decreased with pH, but responses from subjects with high flow rates (HF) differed from those with lower salivary flow (LF). Salivary pH and Na + was higher in HF subjects. In response to change in sample pH and TiA, HR has large changes in SAL, but little change in SI, while the LF showed marked changes in SI with little change in SAL.

                            Impacts
                            (N/A)

                            Publications


                              Progress 01/01/81 to 12/30/81

                              Outputs
                              On sabbatical, "quality" (Q), sensory properties and composition of 24 1976 Bordeaux wines were evaluated. By Canonical Variates Analysis, of intensity ratings of 9 aroma terms and of 5 flavor by mouth (F by M) terms, the major difference among the aromas was shown to be due to the "green bean/green olive" note and in FbyM to variation in bitterness and astringency. Neither configuration clustered the wines by commune of origin, nor was Q rated differently among the wines. In water, at constant pH, minimum differences (Min) of 0.17 and 0.41 (g tartaric acid/l) in titratable acidity (TiA) and total acidity (ToA), respectively were needed to produce detectable sourness differences. When ToA was held constant, Min of 0.10 units of pH and 0.17 TiA were needed. At constant TiA, no correlation was found between either pH or ToA and sourness (S); at constant pH, both TiA and ToA were significantly correlated with S.

                              Impacts
                              (N/A)

                              Publications


                                Progress 01/01/80 to 12/30/80

                                Outputs
                                To evaluate sourness (S) of organic acids, pairs of acids which were equi-pH andof equal concentrations of titratable protons were tested. Citric acid was less sour than fumaric, succinic, malic, tartaric and lactic. With the exception of lactic being more sour than fumaric, the other acids were equisour. Wines made with increasing skin contact time (SCT) were evaluated by descriptive analysis and by instrumental analysis of headspace volatiles (HS) and non-volatiles. Two-year-old Cabernet Sauvignon wines had an increase in "berry" aroma and a decrease in "vegetative" aroma as the SCT was increased from 2 to 3.5 days. No change in intensity was observed as SCT was increased to 7 days. In 1978 wines, only the 2 and 7 day SCT wines could be discriminated from each other. By Principal Component Analysis of the HS data (HS PCA) and of the "non-volatile data" (NV PCA), the biggest difference among the 77 and 78 wines was that due to vintage, the 77 wines, although from less ripe grapes, having lower acidity and higher K uptake. The next largest difference was due to SCT; the NV PCA, this was attributable to the increase in phenols and extract with extended SCT.

                                Impacts
                                (N/A)

                                Publications


                                  Progress 01/01/79 to 12/30/79

                                  Outputs
                                  Intensity of sourness (S) of aqueous solutions of tartaric acid buffered to pH 3.5, and of solutions buffered to different pHs was scored. Unstructured and structured category scales were used to study if the scales were used differently. As expected, samples with higher concentrations of acid or of lower pH were more sour. Pair tests were then used to pinpoint minimum detectable differences. To evaluate the difference in S of organic acids, equi-pH and equal titratable protons (TP) mixtures of 2 acids were evaluated by pair tests. Within each pair, the concentrations of each acid were selected so that the equi-pH and equi-TP criteria were met, with one sample predominantly one acid, and the second, the other acid. In this manner, it was demonstrated that malic and tartaric acids, and lactic and malic acids are equally sour. Lactic acid is more sour than tartaric; tartaric is more sour than citric. Wines made with increasing skin contact time (SCT) were evaluated by descriptive analysis (DA). Volatile headspace data from 36 white wines of six different grape varieties were analyzed by Principal Component analysis (PCA). The first principal component, using 18 GC peaks, clustered the non-Riesling wines together but effectively separated them from the Riesling wines. To continue data collection, Cabernet Sauvignon wines were made with skin contact intervals of 0-7 days. Wine headspace and non-volatile composition is being determined.

                                  Impacts
                                  (N/A)

                                  Publications


                                    Progress 01/01/78 to 12/30/78

                                    Outputs
                                    The system used for gas chromatographic (GC) headspace (HS) analysis was furtherimproved by glassware modifications. The effect on reproducibility on variation in analysis end point was shown to be insignificant. Cabernet Sauvignon wines were prepared with 7 skin contact times (SCT) and their HS volatiles analyzed by GC. Although volatile profiles were observed. Chardonnay wines were prepared from vines located in different soil types, the sites varying primarily in percentage of clay. Soil pHs were highly significantly correlated with must (or wine) total acidity, but not with must (o- wine) pH. No consistent differences between wines from the different soil types within a vineyard were noted in the aromas or flavors. At 30 months, however, wine from the sites with lower clay content were shown to be significantly "more oxidized" than those from the higher clay content site.

                                    Impacts
                                    (N/A)

                                    Publications


                                      Progress 01/01/77 to 12/30/77

                                      Outputs
                                      By multiple regression of gas chromatographic (GC) data and sensory ratings, theintensities of 6 descriptive terms (Vegetative Woody, Raisiny, Artificial fruit, and Varietal) were predicted significantly for 10 Cabernet Sauvignon wines using 2 or 3 GC peaks as the independent variables. To evaluate the aroma changes produced by increasing skin contact times, volatiles of fermenting musts (I) and of the finished wines (II) are being analyzed by GC analysis for the headspace volatiles. In a study of the preception of viscosity and of the effect of wine components on viscosity, the viscosities of several systems have been measured by capillary viscometry and sensory evaluation. The minimum difference in glycerol detectable in wine was 12 g/1, which produced a viscosity change in the base wine of only 0.07 cp. Because of the confounding effect of the taste of glycerol on perception of viscosity, the minimum difference in viscosity detectable in water (0.7 cp) was determined using carbomethoxylcellulose. The minimum difference in viscosity detectable in wine is currently being determined, as are the viscosities of II.

                                      Impacts
                                      (N/A)

                                      Publications


                                        Progress 01/01/76 to 12/30/76

                                        Outputs
                                        Sensory analyses of 1975 mechanically harvested (M) wines vs the hand harvested (H) showed the M to be higher in phenols and in pH, although no significant difference in scores (using the 20 point card) were shown which were attributable to the harvesting method. In pair preference comparisons, of the Mvs H, only for the 6 hr Chenin blanc was a significant preference demonstrated (in this case for the H). (Twenty four hr lots of Chenin blanc, 6 and 24 hr lots of French Colombard showed no significant preference for M or H.) Chardonnay wines from two sites were made with skin contact times of 0, 5, 10 and 16 hrs. For any contact time rating of astringency, bitterness, flavor intensity and fruitiness were not significantly different from the control in one site (although the values increased with contact time). In the second site,bitterness, fruitiness and flavor intensity increased significantly; astringencyshowed no difference. 15 Cabernet Sauvignon wines were analyzed by headspace analysis following a study of the system reproducibility in which coefficients of variation for 3 runs of the peak areas ranged from 0.01 to 0.40. The same 15 wines were analyzed by descriptive analysis. Correlation of these two analyses are in progress at this time. GC variables were reduced by eliminating those which were not

                                        Impacts
                                        (N/A)

                                        Publications


                                          Progress 01/01/75 to 12/30/75

                                          Outputs
                                          Sensory and chemical analyses of hand (H) and mechanically harvested (M) wines and of wines (L) to which leaves were added, were made. With increased contact time and leaf content, total phenols and color increased. Scores for M or L were not systematically lower than for H, and sometimes were higher. The study has been repeated in triplicate. Astringency (A) and bitterness (B) of incremental amts. of total tannin (T) in a model system and of 4 tannin fractions in white wine were evaluated. As T was raised, B increased, and then decrease, while A increased continually. All 4 fractions were characterized as being both A and B. With increased amts. of catechins, "small" and "large" anthocyanogens, A changed sig., changes in neither A in condensed tannis (CT) nor B in any of the 4 were not sig. Per unit weight, CT was the most A and most B. Gas chromatrographic (GC) results of volatile analyses of 4 Riesling (R) and5 non-riesling (NR) wines were analyzed by discriminat analysis (DA) and multiple regression analyses (MDA). Prediction of the intensity of the R aroma by MDA or sorting of the wines into R and NR categories by DA was usually possible in one step. Using GC profiles of other but similar wines generated different equations, using different peaks as variables, which were predictive within new sets. More samples need to be included in the analyses for more generally predictive results.

                                          Impacts
                                          (N/A)

                                          Publications


                                            Progress 01/01/74 to 12/30/74

                                            Outputs
                                            In must and model must systems, the intensity of "leafiness" increased with increasing percentage of leaves added, increasing leaf contact time and/or temperature. Wines have been made from 4 grape varieties which were mechanically harvested (MH) and hand harvested (HH). In preliminary evaluation of the wines including HH to which 1, 3 and 5% leaves had been added, no major defects attributable to leaf contact have been detected. Must pH generally increased in lots held 24 hr before crushing. Panelists are being trained to detect bitterness and/or astringency in wine systems. Taste qualities of (total) tannin at levels usually found in wine are being assessed. In model wine systems, for total tannin, the intensity of bitterness is not directly proportional to concentration, although astringency approximately is. For further investigation: wine was made from grapes (Cabernet Sauvignon) fermented on the skins 0-9 days (for future volatile analysis), wine has been made from 4 grape varieties from hot and cool regions for sensory and volatile analysis to examine components particularly important in "varietal" aromas.

                                            Impacts
                                            (N/A)

                                            Publications