THE USE OF SENSORY METHODOLOGY TO STUDY THE EFFECTS OF FRUIT MATURITY ON GRAPES & WINES; FOOD & WINE-PAIRINGS; MOUTH FEEL OF WHITE & RED WINES; VARIETY & DRYING ON RAISINS.

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 1000616
• Project Start Date: Oct 31, 2013
• Project End Date: Sep 30, 2018
• Program Code: [(N/A)]- (N/A)

Recipient Organization
UNIVERSITY OF CALIFORNIA, DAVIS
410 MRAK HALL
DAVIS,CA 95616-8671

Performing Department
Viticulture and Enology

Non Technical Summary
Both the wine and the almond project relates to quality and consumer perceptions of quality. Generally, quality is related to satisfaction and occurs when the perception of the performance of the product meets the user's prior expectations of that product. These prior expectations of quality are based on extrinsic (brand, packaging, etc.) and intrinsic (flavor type and intensity) factors. However, if during product consumption, the intrinsic factors do not match the consumers' perceptions of the extrinsic factors then the consumers may be disappointed and may not purchase the product again. A complete study of the evaluation of quality including both extrinsic and intrinsic factors would be ideal but it would be very large and expensive. Starting with only intrinsic factors is a sensible way to make the project manageable, especially since a great deal of especially wine quality discussions are based on blind tastings i.e. Wine Spectator, Parker Quality scores, Wine Fairs, etc. In Sensory Science, the perceived intrinsic quality of a product is described variously. One school of thought is that "quality is the absence of defects". In this case the "lack of defect" is likely to be true for most commercially produced products yet there are clear differences between the sensory attributes of a simple inexpensive non-aged red wine and a complex oak-aged red wine. In some sense, these differences in the complexity of the sensory attributes, as perceived by a trained descriptive panel, should translate into a difference of perceived quality; yet the definition that quality is a lack of defects does not allow that. The other school says, "Quality may only be evaluated by experts who have had long term experience with a wide range of the qualities of the products within the category." Anyone else, asked to rate quality, will score the product in terms of their subjective liking. Here, the problem becomes finding and evaluating these "experts" that have had long term experience with both defective and non-defective products. In the case of coffee and tea tasters, there are apprenticeships within these industries to educate the quality evaluators in terms of the qualities found in the product category. Due to this extensive training, it is likely that the quality evaluation of successive tea tasters would be quite similar. In the wine industry, for example, for wine competitions, the experts are frequently wine writers, winemakers, wine educators or chefs. These experts would not have had a similar 'apprenticeship' in evaluating wine quality and their judgments could be very dissimilar. The wine proposal will compare wine consumers and experts in terms of their similarity, hedonic and quality scores. Additionally, the use of three different wine quality score cards by wine experts will be compared. Additionally, all these data will be integrated with the results of descriptive analysis of the wines by trained panelists as well as with volatile and non-volatile chemical data. In the almond proposal we essentially propose the exact same thing with one exception - there are no 'expert' almond judges.

Animal Health Component

70%

Research Effort Categories

Basic

20%

Applied

70%

Developmental

10%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
502	1131	3090	80%
503	1212	3090	20%

Knowledge Area
503 - Quality Maintenance in Storing and Marketing Food Products; 502 - New and Improved Food Products;

Subject Of Investigation
1212 - Almond; 1131 - Wine grapes;

Field Of Science
3090 - Sensory science (human senses);

Keywords

quality

wine

expert judgments

almonds

consumer liking

Goals / Objectives
The overall objective is to determine if quality can be reliably evaluated by wine consumers, wine aficionados and/or wine experts and what their quality scores actually mean relative to the intensities of the sensory attributes we get from trained panels, the volatile and non-volatile chemical data, and the liking scores we get from wine consumers. The specific objectives are[1]: To compare the perceived liking scores also known as hedonic scores of wine experts, wine aficionados and wine consumers within each group with their quality scores. To compare the similarity maps and the quality scores of the three groups. To compare the results from the three groups with descriptive analysis data of the wines obtained from a trained objective panel. To compare the descriptive data, the quality data from the three groups, the results from the score sheets to the volatile and non-volatile chemical data. Almond storage is dependent upon many factors including the development of rancidity. To date, little is known regarding correlations between chemical markers of rancidity and consumer acceptance/perception. In recent work, we identified several chemical markers (i.e. certain oxiranes, carbonyls, and short chain acids), which are absent in raw and freshly roasted almonds, but detectable after a short period of storage. These compounds reflect advancing stages of oxidation and are independent of processing or storage history. Additionally, volatiles including hexanoic acid, octanal, [E]-2-octenal, heptanol, and octanol appeared earlier in stored almonds than more traditional markers of rancidity (i.e. hexenal and nonenal). Significant decreases in roasting-related aroma volatiles were observed by ten weeks of storage. Based upon these observations, we propose to study the relationship between the development of these compounds, the sensory attributes of the almonds, and consumer hedonic responses these chemical changes. The primary objectives are to determine whether it is possible to use chemical or sensory markers to predict the consumer hedonic responses to almond quality. Perform a sensory descriptive analysis of the 12 treatments at least monthly (Lawless and Heymann, 2010). This will allow us to determine which sensory attributes change during the storage period. Determine the volatile profile of the 12 treatments at least monthly. This will allow us to determine which volatile compounds change during the storage period and how roasting and temperature impact these changes. Determine the consumer hedonic response (degree of liking) for the 12 treatments monthly (Lawless and Heymann, 2010) at least monthly. This will allow us to determine whether the chemical and sensory changes found in objectives 1 and 2 impact consumer liking. Compare the sensory and chemical changes to the consumer hedonic information on a monthly basis and over the course of the 12 month study period. [1] Specific objectives 1 and 2 were funded in the 2012 to 2013 cycle.

Project Methods
Experimental Procedures to accomplish Objective(s): Procedures for the wine quality study: Panelists: To meet objectives 1 and 2 three groups of panelists (about 24 per group) will be chosen. The selection criteria for the wine experts (WE-I) are that they are employed in the wine industry, that they formally taste wine at least 3 times a week and that their wine evaluation results are used to make business decisions. The selection criteria for the wine aficionados (WA) are that they taste at least 3 times per week and that they have been actively involved in learning about winemaking. The selection criteria for the wine consumers (WC) are that they drink wine once a week and they are not affiliated with the wine industry. To meet objectives 3 and 4, a trained panel of 8-12 panelists (DA) will evaluate the wines using standard descriptive analysis methodology. Wines: The wines chosen for this study will come from California and they will all be the same variety. We will use actual quality scores from an independent wine judging event to select our wines Wine evaluations: To meet objectives 1 and 2, the WE-I, WA and WC groups will be asked to do the following: Evaluate the 27 wines in randomized order in coded clear wine glasses and indicate their perceived liking for the wines using the standard 9-point hedonic scale. Sort the 27 wines, presented in randomized order in clear wine glasses with different codes, into similar groups. Panelists will be instructed that they may use as few as 2 and as many as 26 similarity groups. Sort the 27 wines, presented in randomized order in clear wine glasses with a third set of different codes, into quality groups. To meet objectives 3 and 4, a panel comprising 10 to 16 judges will be trained by the consensus method to describe the wines using specified sensory attributes. Attributes will be anchored with reference standards. All wines will be evaluated in triplicate. Wines will be served to panelists using a completely randomized design in clear wine glasses. To meet objective 4, the 27 wines will be profiled in a non-targeted way for volatiles with gas chromatography-mass spectrometry (GC-MS), non-volatiles with high pressure liquid chromatography-mass spectrometry (HPLC-MS) and macro-and micro-elemental composition using inductively coupled plasma with both mass spectrometry (ICP-MS) and optical emission spectroscopy (ICP-OES). Data Analyses: Objective 1: Hedonic scores will be analyzed by analysis of variance with main effects: wine, group (WC, CA and WE-I) and the wine by group interaction. This would indicate whether the groups liked the wines in similar ways or not. Additionally an internal preference map of the data would indicate in a two- or three-dimensional map how the groups did or did not differ in their liking of the wines. Objective 2: Similarity scores will be analyzed by the creation of a multidimensional map using DISTATIS (Abdi et al. 2007). This would indicate if the wines falling in the different price categories were perceived to be similar and what effect, if any, the addition of defects had on these perceived similarities. Objective 3: The descriptive analysis data of the trained DA panel will be analyzed by analysis of variance (main effects: wine, panelist, replication; and all two-way interactions). These data will also be evaluated by a multivariate technique such as principal component analysis (PCA) or canonical variate analysis (CVA). Both these techniques will create a two or three-dimensional map of the wine space. The PCA or CVA space will be compared to the DISTATIS similarity space to determine the latent dimensions used by groups WC, WA and WE-I to determine the wine similarities. The PCA or CVA space and the hedonic scores of groups WC, WA and WE-I as well as the quality scores of these three groups and those on the three score sheets by groups WE-II will be used to create external preference and quality maps to compare perceived liking and quality across groups. Objective 4: The volatile and non-volatile chemical data will be analyzed by analysis of variance (main effects: wine, replication). These data will also be evaluated by a multivariate technique such as principal component analysis (PCA) or canonical variate analysis (CVA). Both these techniques will create a two or three-dimensional map of the wine space. The PCA or CVA space will be compared to the DISTATIS similarity space to determine the latent dimensions used by groups WC, WA and WE-I and the DA panel to determine the wine similarities. Procedures for the almond quality project Objective 1: We will train 15 panelists[1] (Heymann, et al. 2012) to describe the differences among different almond varieties and almonds that have differing storage conditions. The panelists, through a consensus training process, will determine the sensory attributes that will be evaluated throughout the storage period. Additionally, they will agree on the reference standards to anchor these attributes. Each panelist will then evaluate the whole almonds monthly in triplicate. The data would be analyzed by analysis of variance and canonical variates analysis. Almond volatiles will be analyzed using methodology previously developed on our campus (Mitchell 2010). Briefly, volatiles will be extracted using solid phase micro-extraction (SPME). A mixed internal standard solution will be used for quantization. SPME fibers will be allowed to equilibrate with samples at room temperature for 30 minutes. A GC system coupled to a mass selective detector will be used to identify and quantify volatiles in almond samples. Total ion chromatograms will be collected scanning from m/z 30 to 150 at a rate of 3.06 scans/sec. The data collected will be analyzed by analysis of variance. Objective 3: At each time point at least 100 almond consumers[2] will evaluate the 12 treatments. The consumers will be asked to indicate their like/dislike for the almonds on the standard 9-point hedonic scale. We will also ask some demographic questions of the consumers each month. The data would be analyzed by analysis of variance, cluster analysis and external preference mapping. Objective 4: Compare the sensory and chemical changes to the consumer hedonic information on a monthly basis and over the course of the 12 month study period. For this purpose we would use external preference, multiple factor analysis, and various regression analyses. [1] This is more than the usual 8 to 12 but we will likely have a few not be available during some of the months. This way we should have at least 10 at each month's evaluations [2] Screened to consume almonds at least monthly

Progress 10/31/13 to 09/30/18

Outputs
Target Audience:The target audience for this project are sensory scientists, both in the wine industry as well as in the food industry. Additionally we also target winemakers, mead makers and brewers. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Additionally, during this time I trained six post-doctoral fellows, three PhD candidates and four MS students. I also employed about 15 undergraduate students as laboratory employees. How have the results been disseminated to communities of interest?Through publications, short courses and presentations at National and International meetings. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? During the period of this project approximately 38 refereed publications were published, two book chapters and one book. Additionally, information from this project was used in about 15 presentations in California, the USA and in other countries.

Publications

• Type: Journal Articles Status: Published Year Published: 2017 Citation: 1. Watrelot, A.A., Byrnes, N.K., Heymann, H., and Kennedy, J.A. (2017) Understanding the relationship between red wine matrix, tannin activity and sensory perception. Journal of Agriculture and Food Chemistry, 64: 9116-9123.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: 1. Frost, S.C., Harbertson, J., and Heymann, H. (2017). A full factorial study on the effect of tannins, acidity, and ethanol on the temporal perception of taste and mouthfeel in red wine. Food Quality and Preference, 62: 1-7.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: 1. Murray, N.M., ORiordan, D.O., Jacquier, J-C., OSullivan, M., Choen, J.L., Heymann, H., Barile, D., and Dallas, D.C. (2017). Validation of a paper-disk approach to facilitate the sensory evaluation of bitterness in dairy protein hydrolysates from a newly developed food-grade fractionation system. Journal of Sensory Studies, 32 (3) e12266.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: 1. Sherman, E., Greenwood, D.R., Villas-Boâs, S.G., Heymann, H., and Harbertson, J.F. (2017). Impact of Grape Maturity and Ethanol Concentration on Sensory Properties of Washington State Merlot Wines. American Journal of Enology and Viticulture, 68: 344-356.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: 1. Johnson, A.J., Hopfer, H., Heymann, H. and Ebeler, S.E. (2017). Aroma Perception and Chemistry of Bitters in Whiskey Matrices: Modeling the Old-Fashioned. Chemosensory Perception, 10 (4), 135-148.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: 1. Sherman, E., Harbertson, J.F., Greenwood, D.R., Villas-Bôas, S.G., Fiehn, O., and Heymann, H. (2017). Reference samples guide variable selection for correlation of wine sensory and volatile profiling data. Food Chemistry 267:344-354.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: 1. Lestringant, P., Delarue, J., and Heymann, H. (2017). Do panelists memorize products when performing descriptive analysis on few products? Journal of Sensory Studies, 33 (1), e12305.
• Type: Journal Articles Status: Published Year Published: 2018 Citation: 1. Frost, S.C., Blackman, J., Ebeler, S.E., and Heymann, H. (2018). Analysis of temporal dominance of sensation data using correspondence analysis on Merlot wine with differing maceration and cap management regimes. Food Quality and Preference, 64:245-252.
• Type: Journal Articles Status: Published Year Published: 2018 Citation: 1. Lahne, J., Abdi, H. and Heymann, H. (2018). Rapid sensory profiles with DISTATIS and Barycentric Text Projection: An example with amari, bitter herbal liqueurs. Food Quality and Preference, 66: 36-43.
• Type: Journal Articles Status: Published Year Published: 2018 Citation: 1. Lestringant, P., Delarue, J., and Heymann, H. (2018). Effects of adding extra samples to a product set when using descriptive analysis, Food Quality and Preference, 66: 127-140.
• Type: Journal Articles Status: Published Year Published: 2018 Citation: 1. Liu, J., Wender, B., Sherman, E., Harbertson, J.F., and Heymann, H. (2018). Comparison of rapid descriptive sensory methodologies for evaluating red wines. Food Research International , 106, 892-900.

Progress 10/01/16 to 09/30/17

Outputs
Target Audience:My target audiences are winemakers, sensory practitioners in industry as well as sensory scientists and wine researchers in California, nationally and internationally. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?During this period I had visiting scholars from Japan and Argentina. How have the results been disseminated to communities of interest?Work from this project was presented at the Rosemary Pangborn Memorial Symposium. What do you plan to do during the next reporting period to accomplish the goals?Continue working on the various aspects of the project.

Impacts
What was accomplished under these goals? Descriptive analysis (DA) has been widely applied to describe the sensory characteristics of products. An analysis of the scientific literature was performed to better understand the context in which descriptive methods are utilized and how they relate to newly developed rapid methods. How representative of the application of DA are the experiments aiming at developing methods published in sensory journals? Web of Science was queried using "sensory profile" and "descriptive analysis" as search terms from years 2010 to 2015. For each generated result, the following metadata were extracted: number and type of samples, number of panellists, training and evaluation methods, and data analysis. The data were analysed according to the goals of the study (using or developing sensory methods) and type of journal publishing the article (sensory specific journal or other). Out of 389 entries, 351 used DA, 26 used DA in conjunction with other descriptive methods, and 38 only used rapid methods. The most used rapid methods were Napping / Projective Mapping and Sorting. Since 2012, the number of experiments using rapid methods has slightly increased while DA applications have decreased. Rapid methods were primarily published in sensory journals, in articles aimed at developing new methods. In contrast, 91% of DA experiments were used to obtain information about products in non-sensory specific journals. Alcoholic beverages, produce, dairy and non-alcoholic beverages were the top 4 product categories assessed. However, produce was underrepresented in studies published in sensory journals. The range of the number of products used was wider for studies using descriptive methods, with more experiments using very few (less than 4) or very many (more than 12) samples. Some results suggest a discrepancy in context between experiments designed to investigate sensory methods or to obtain sensory information. Context of use should be considered when testing new methods. Applying descriptive analysis to small sample sets can be problematic, due to memorization effects or small sample statistics. Additional samples can be added to account for these issues. Yet, adding samples modifies the evaluation context and could result in differences in descriptions of the samples. This study investigated the effect of adding samples, depending on how many samples are added to the target set and how they are chosen. In two separate sets of experiments, panels were trained using a target set of four samples. They evaluated the target set, and then, two sets with added samples. To evaluate the effect of the type of samples added, two products were added, which were either similar to the target samples or different. To investigate the effect of the number of products, either four or six samples were added to the target. Each experiment was performed on two product categories, ketchup and lemonade, to assess the reproducibility of the results. Multi- and Univariate Analysis of Variance, and Canonical Variate Analysis (CVA) were performed on each data set, keeping only the four target samples. Following that, a Multiple Factor Analysis was performed on all data sets. Overall, the product structure was well maintained in spite of different evaluation contexts. Differences among treatments were within the margin of those found between the two panels conducting the same part of the study. However, separate analyses showed that the complexity of each product space increased with the number of samples, as indicated by higher variance on dimensions 2 and 3 of the CVA. Moreover, the most salient descriptors on the first two dimensions depended on the evaluation context. This study showed that it is possible to add samples to a target set. However, the reseacher should be aware of the possible impact on product descriptions.

Publications

• Type: Books Status: Published Year Published: 2016 Citation: Heymann, H., and Ebeler, S.E. (2016). Sensory and Instrumental Evaluation of Alcoholic Beverages. Elsevier, New York, NY.
• Type: Journal Articles Status: Published Year Published: 2016 Citation: Waterhouse, A.L., Frost, S.C., Ugliano, M., Cantu, A.R., Currie, B.L., Anderson, M., Chassy, A.W., Vidal, S., Di�val, J-B., Aagaard, O., and Heymann, H. (2016). Sulfur Dioxide - Oxygen Consumption Ratio Reveals Differences in Bottled Wine. American Journal of Enology and Viticulture, 67: 449-459
• Type: Journal Articles Status: Published Year Published: 2016 Citation: S.C. Frost, J.W. Blackman, S.E. Ebeler, H. Heymann. Analysis of Temporal Dominance of Sensation data using correspondence analysis on Merlot wine with differing maceration and cap management regimes Food Quality and Preference (2016), 10.1016/j.foodqual.2016.11.011

Progress 10/01/15 to 09/30/16

Outputs
Target Audience:My target audiences are winemakers, sensory practitioners in industry as well as sensory scientists and wine researchers in California, nationally and internationally. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?During this period I had visiting scholars from New Zealand, Denmark, Costa Rica via The Netherlands, Ireland and Austria. I also had two post-doctoral fellows, one a US citizen and one a German citizen. How have the results been disseminated to communities of interest?Work from this project was presented at the American Society of Enology and Viticulture Annual Meeting; at the Sensometrics conference and at the EuroSense conference. What do you plan to do during the next reporting period to accomplish the goals?Continue working on the various aspects of the project.

Impacts
What was accomplished under these goals? Interest in and consumption of rosé wines has increased considerably worldwide in recent years, and that trend continues to be on the rise. The allure of rosé is driven by its unique sensory profile and chemical composition, which are arrived at through certain vinification and viticultural practices. Therefore, insight into which production methods contribute to specific sensory qualities carries a high value in this burgeoning market. This study provides the first comprehensive interpretation of the relationship between the three most common rosé production techniques and their corresponding sensory and chemical properties. Two typical rosé cultivars, Syrah and Tempranillo, underwent each of the three methods, short maceration, saignée and blending, to produce six rosé wines. Trained panelists through Qualitative Descriptive Analysis (QDA) and Temporal Dominance of Sensation (TDS) determined sensory profiles for each wine. Effects of color perception on attribute assignment were also explored. Chemical analysis was carried out on each wine and results were correlated to sensory data. All evaluations were performed in triplicate. Cap management and maceration describe techniques that manipulate the contact between juice and grape solids during winemaking. These practices manipulate the extraction of phenolic compounds with the intent of altering taste and mouth-feel. The effect of four cap management techniques, along with five post fermentation maceration time points was evaluated for phenolic profile and sensory. All experiments were conducted in triplicate using Merlot grapes, with 9 different treatments. Fruit was crushed into 27 fermentors (200 L). A three times daily pump-over regime (one juice volume) was applied to 18 of the 27 fermentors. At dryness, three of the 18 fermentors were pressed, and the remaining 15 fermentors were pressed in groups of three at intervals of 1, 2, 4, 6 and 8 weeks. Six of the remaining nine fermentors were fitted with a purpose built wire mesh to submerge the cap into the fermentation. At dryness three were pressed and three underwent an eight week maceration then pressed. The final three fermentors were punched down three times daily then pressed at dry. The changes induced by the different winemaking treatments were characterized with instrumental and sensory methods. The wines sensory profiles were determined with descriptive analysis (DA) and temporal dominance of sensation (TDS). Hydrophilic interaction chromatography (HILIC) coupled with quadrupole time-of-flight mass detection (qTOF) was used to evaluate the chemical distribution of tannin species. Baseline resolution of proanthocyanidin isomers was achieved, allowing for relationships between polymerization and winemaking technique to be observed. Multivariate analysis of the qTOF polyphenol measurements discriminated the 9 treatmensts. Longer maceration times produced proportional increases in specific polyphenol compounds. Extraction patterns and sensory implications will be reported for specific compounds and compound classes. Rapid Methods (RM) has been considered as an interesting alternative for defining sensory characteristics of food products. These methods are less time consuming and more flexible compared to more traditional methodologies like Descriptive Analysis (DA). The aim of this study was to determine the sensory profile of a set of six wines with none (control), one, two, four, six and eight weeks of extended maceration (EM) applying three Rapid Methods, Flash Profile (FP), Sorted Napping (SN) and Free Sorting (FS). The three RM results were compared with DA results from a previous study carried out with the same set of wines. For the three techniques all wine samples were presented at the same time to the assessors and trials were done in duplicate. A blind control was included to evaluate individual assessors' performance based on the People Performance Index (PPI). Fifteen assessors with previous experience in sensory evaluation participated in each of the techniques trials (N=45). Results showed there were differences among the RM; this influenced the product space, the attributes definition and the assessors' performance. FP presented the most consistent results, wines with lower EM time were perceived similar. FS and SN leaded to more global results, with both techniques was more difficult to see an effect of EM time, only two wines were distinctively identified (2 and 8 weeks). Individual configurations for the products plots, sensory attributes and RV coefficient were used to compare RM with DA. FP presented more similarities with DA (RV=0.59). Previous studies consider attribute-based techniques like FP leaded to more comparable results with DA than holistic techniques. There were significant differences (p<0.05) between PPI values calculated with the three RM. FP got the best values and the assessors' performance was more consistent from one repetition to the other. FS leaded to the worst PPI values.

Publications

• Type: Journal Articles Status: Published Year Published: 2015 Citation: Blanco-Ulate, B., Amrine, K.C.H., Collins, T.S., Rivero, R.M., Vicente, A.R., Morales-Cruz, A., Doyle, C.L., Ye, Z., Heymann, H., Ebeler, S.E., and Cantu, D. (2015). Developmental and metabolic plasticity of white-skinned grape berries in response to Botrytis cinerea during noble rot. Plant Physiology, 169: 2422-2443
• Type: Journal Articles Status: Published Year Published: 2016 Citation: Hendrickson, D.L., Lerno, L.A., Hjelmeland, A.K., Ebeler, S.E., Heymann, H., Hopfer, H., Block. K.L., Brenneman, C.E., and Oberholster, A. (2016). Impact of mechanical harvesting and optical berry sorting on grape and wine composition. American Journal of Enology and Viticulture DOI: 10.5344/ajev.2016.14132
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Lahne, J., Collins, T.S. and Heymann, H. (2015) Replication improves Sorting-Task results analyzed by DISATIS in a consumer study of American Bourbon and Rye Whiskeys. Journal of Food Science 81: S1263-S1271

Progress 10/01/14 to 09/30/15

Outputs
Target Audience:My target audiences are winemakers, sensory practitioners in industry as well as sensory scientists and wine researchers in California, nationally and internationally. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The project has had two post-doctoral fellows, one PhD student and visiting scholars from the USA, Australia and New Zealand working on various aspects. How have the results been disseminated to communities of interest?Work from this project was presented at the American Society of Enology and Viticulture Annual Meeting; at the Metabolomics Conference, both in the USA, and at In Vino Analytical Scientia Meeting in Italy. What do you plan to do during the next reporting period to accomplish the goals?Continue working on the various aspects of the project.

Impacts
What was accomplished under these goals? Maturity is known to have significant impacts on the non-volatile and volatile components in wine grapes, particularly during the exponential accumulation of soluble solids. Ethanol has numerous impacts on wine chemistry of sensory importance including impacting the extraction of many non-volatile compounds during maceration and the partitioning of volatile compounds between the vapor and liquid phase in finished wine. As there is a direct relationship between the resulting ethanol concentration and the initial soluble solids of a must we sought to evaluate the interaction between ethanol and grape maturity during the crucial final stages of grape ripening. To this endMerlot grapes from Washington State were harvested at three different maturities: ~20 Brix, ~24 Brix and ~28 Brix. Chaptalization or saignée followed by water-addition were used to target different ethanol concentrations at each harvest while maintaining the juice to skin ratios originally present in the must. Portions of each harvest were chaptalized or saignée followed by water-addition to obtain nine treatments: three harvest dates at three soluble solids levels, which were fermented in triplicate. A trained descriptive panel (N=14) rated the intensity of 16 aroma, 15 in-mouth flavor, 3 taste, and 3 mouth-feel attributes using a line scale. The panel differentiated the wines using attributes characteristic of green notes, including `herbaceous' and `vegetal' as well as attributes characterizing fruit notes, including `dark fruit', `red fruit', and `dried fruit'. The volatile and non-volatile composition of the wines was analyzed using a metabolomics approach (untargeted chemical analysis of biological systems). Multivariate analyses of peak heights from gas and liquid chromatography - mass spectrometry experiments showed that features detected by untargeted mass spectral analysis clearly separated the treatments in both the volatile and non-volatile profiles. American whiskeys are some of the most important distilled beverages in today's market; in particular, "bourbon" (mashbill more than 50% corn) and "rye" (mashbill more than 50% rye) are highly sought-after and are generally thought to be very different products. In contrast to their economic and cultural importance, there is almost no scientific information about the sensory properties of these products. In this study, 10 American whiskeys - 5 bourbons and 5 ryes - were sorted into groups by 21 consumers over 2 replications. The grouping data were analyzed using several statistical approaches: Multidimensional Scaling (MDS), standard and replicated DISTATIS, and Hierarchical Cluster Analysis (HCA). The relationship between mashbill, age, producer, and consumer perceptions of similarity in the whiskeys were examined, as was the usefulness of replicated DISTATIS compared to less novel approaches (MDS and HCA). Replicated DISTATIS was found to provide somewhat more stable, detailed results than standard DISTATIS, MDS, or HCA, although all analyses of data produced roughly similar results. In all of the analyses, no significant relationship between mashbill and consumer perceptions of similarity in whiskey was found; in contrast, there appeared to be strong evidence of producer and aging effects. Thus, this research provides the first published evidence that mashbill is not a strong driver of the sensory properties of American whiskey: bourbon and rye whiskeys, while legally distinct, may not be distinguishable on the basis of their sensory properties. Maceration describes a process whereby grape solid material is in contact with juice or wine. During winemaking, maceration allows for phenolic compounds to migrate into the juice/ wine from the skins and seeds. Post fermentation maceration (extended maceration) is intended to allow differential tannin extraction after the alcoholic fermentation, thus modifying the tannin and sensory profile[1]. We evaluated the effect of maceration length on perceived sensory and measured chemistry. Merlot grapes were crushed into 18 200 L fermentation vessels. During fermentation, 1 volume of juice was pumped over the cap 3 times daily. Three of the 18 fermentors were pressed at dryness as the control treatments. The remaining fermentors were allowed to continue maceration and then pressed in triplicate groups at the following intervals; 1, 2, 4, 6 and 8 weeks. During the maceration period each fermentor was pumped over for 5 minutes once per day. Descriptive analysis (DA) and temporal dominance of sensation (TDS) described the sensory profile of the wines. The sensory profiles were then related to malvidin-3-glucoside (m-3-g), protein perceptible tannin, polymeric pigment contents, tannin species distribution (LC-MS/MS), and volatile profiling (GC-MS). Aromas of red fruit and citrus/ floral increased with up to 6 weeks of maceration. Bitterness, astringency, drying, and hotness also increased with maceration length. The TDS results showed similar profile for each wine. The wines were initially sweet/sour, followed by a brief astringency that receded to hotness. The short period of hotness ended as the astringency reemerged. The time duration of the astringent finish was related to increased maceration. Treatments of 6 and 8 weeks of maceration were unique in that they finished with bitterness as the primary sensation. Increased tannin concentrations were measured between the control and 1 week of maceration, but no differences were found among treatments of 1 to 6 weeks. Week 8 had significantly higher tannin concentrations than all other treatments. Maceration length was also associated with decreasing m-3-g and polymeric pigment. The observed decrease in pigment concentration is likely due to adsorption of anthocyanin to grape pomace. The study allowed us to determine compositional and sensory differences due to extended maceration. This scientific data provides an insight into a winemaking practice that has relied largely on anecdotal evidence.

Publications

• Type: Journal Articles Status: Published Year Published: 2014 Citation: Hopfer, H., Nelson, J., Collins, T., Heymann, H., and Ebeler, S.E. (2014) Elemental Profiling of Red Wine - The combined effects of vineyard origin and processing winery. Food Chemistry, Food chemistry 172, 486-496.
• Type: Journal Articles Status: Published Year Published: 2014 Citation: Umali, A.P., Ghanem, E., Hopfer, H., Hussain, A., Kao, Y., Zabanal, A.G., Wilkins, B.J., Hobza, C., Quach, D.K., Fredell, M., Heymann, H., and Anslyn, E. (2014) Grape and Wine Sensory Attributes Correlate with Pattern-Based Discrimination of Cabernet Sauvignon Wines by a Peptidic Sensor Array. Tetrahedron, 71:3095-3099.
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Oberholster, A., Elmendorf, B.L., Lerno, L.A., King, E.S., Heymann, H., Brenneman, C.A. and Boulton, R.B. (2015) Barrel maturation, oak alternatives and micro-oxygenation: Influence on red wine aging and quality. Food Chemistry, 173:1250-1258.
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Hood White, M., and Heymann, H. (2015). Assessing the Sensory Profiles of Sparkling Wine over Time. American Journal of Enology and Viticulture, 66:156-163.
• Type: Journal Articles Status: Published Year Published: 2014 Citation: Cramer, G.C., Ghan, R., Schlauch, K.A., Tillet, R.L., Heymann, H., Ferrarini, A., Delledonne, M., Zenoni, S., and Pezzotti, M. (2014) Transcriptomic analysis of the late stages of grapevine (Vitis vinifera cv. Cabernet Sauvignon) berry ripening indicates significant induction of ethylene signaling and flavor pathways in the skin. BMC Biology, 14:370.
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Johnson, A., Heymann, H., and Ebeler, S.E. (2015). Volatile and sensory profiling of cocktail bitters. Food Chemisty, 179:343-354
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Ghanem, E., Hopfer, H. Navarro, A., Ritzer, M., Mahmood, L., Fredell, M., Cubley, A., Bolen, J., Fattah, R., Teasdalea, Lieu, L., Chua, T., Kim, D., Marini, F., Heymann, H. and Anslyn, E.V. Predicting the Composition of Red Wine Blends using an Array of Multicomponent Peptide-Based Sensors. Molecules, 20:9170-9182.
• Type: Book Chapters Status: Published Year Published: 2014 Citation: Heymann, H. and Hopfer, H. (2014). Ann Noble, Wine and Sensometrics. Proceedings of Evaluating Wine Flavor through Chemistry and Sensory Analyses (Thorngate, J.T., Ed). American Society of Enology and Viticulture, pp.5-11.
• Type: Book Chapters Status: Published Year Published: 2015 Citation: Heymann, H., Robinson, A.L., Stoumen, M.E., King, E.S., Hopfer, H., Boulton, R.B., and Ebeler, S.E. (2015). Ch. 7. Effect of Region on the volatile composition and sensory profiles of Malbec and Cabernet Sauvignon wines. Advances in Wine Research (S.E. Ebeler, G. Sacks, S. Vidal and P. Winterhalter, Eds). ACS Symposium Series 1203.
• Type: Journal Articles Status: Published Year Published: 2015 Citation: Hopfer, H., Nelson, J., Ebeler, S.E. and Heymann, H. (2015). Correlating Wine Quality Indicators to Chemical and Sensory Measurement. Molecules, 20:8453-8483.

Progress 10/31/13 to 09/30/14

Outputs
Target Audience: Wine industry stake holders, sensory science researchers and practitioners Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? I have graduated two MS candidates and have extensively published with two post-doctoral fellows during this time. I also have two PhD students working in the area. How have the results been disseminated to communities of interest? I and coworkers in my laboratory spoke at numerous scientific meetings as well as at outreach events. What do you plan to do during the next reporting period to accomplish the goals? Continue the work we are currently doing.

Impacts
What was accomplished under these goals? Sparkling wine sensory properties are driven by many factors, including the amount of CO2 and pressure in the bottle. We were interested in characterizing the sensory attributes of Californian blanc de blanc sparkling wine, as well as the sensory impact of various effervescence conditions such as artificial nucleation points on the aroma and flavor. Preliminary, the effect effervescence conditions on the aroma and flavor intensity was studied. For this, three different nucleation treatments (air-dried glasses, paper towel dried glasses and glasses with an etched nucleation point) were compared. Thirty-three panelists assessed overall aroma and flavor intensity for each treatment in triplicate. No statistically significant differences in aroma and flavor intensity were found. Secondarily, the sensory profiles of eight Californian blanc de blanc sparkling wines were determined using a generic descriptive analysis (DA). Eleven trained judges used visual, aroma, taste, mouthfeel and bubble descriptors to characterize the sensory differences among the sparkling wines. A potential temporal effect on the aroma and flavor development of Californian blanc de blanc was determined, by performing the DA 1 minute and 5 minutes after pouring, using the same descriptors. In addition, atmospheric pressure of each wine bottle was measured to determine the influence of pressure on taste and aroma over time. Sparkling wines assessed at one and five minutes showed a significant difference in all wines, with regard to time. This difference was characterized by a decrease in bubble attributes and allowed for greater differences within the sample set; thus, at five minutes there was a greater differentiation among the wines. No correlations between bottle pressure and sensory attributes were found. Minerality is a way for wine writers to associate wines with their terroir. Little research has been done on the concept of minerality. In this study white wines were compared by projective mapping (PM) performed by wine industry professionals to a standard sensory descriptive analysis (DA) by trained judges. The PM found minerality to be positively correlated with acid taste and citrus, fresh, wet stone, and chemical aromas, and negatively correlated to butter, butterscotch, vanilla, and oak aromas. The PM panel minerality was associated with both aroma and taste perception. The DA found minerality to be positively correlated with reduced, chalky, and grassy aromas and bitter taste, and negatively correlated with barrel, caramel, honey, juicy fruit, musty and cat pee aromas. Wine groupings were similar between the two panels. Minerality was highly associated with malic acid, tartaric acid and titratable acidity, and moderately associated with free and total sulfur dioxide. Practical Applications There were similar concepts for minerality between the two types of panels but there were also differences. This gives us the hope that teasing out the exact concept of minerality in white wines should be possible.

Publications

• Type: Journal Articles Status: Published Year Published: 2013 Citation: Hjelmeland, A. K., King, E. S., Heymann, H., and Ebeler, S. E. (2013). Characterizing the chemical and sensory profiles of US Cabernet Sauvignon wines. American Journal of Enology and Viticulture 64:169-179
• Type: Journal Articles Status: Published Year Published: 2013 Citation: Jaeger, S., Mielby, L., Heymann, H., Jia, Y.; Frost, M (2013) Analysing conjoint data with OLS and PLS regression: a case study with wine. Journal of the Science of Food and Agriculture 93: 3682 -3690
• Type: Journal Articles Status: Published Year Published: 2013 Citation: Hopfer, H., Nelson, J., Mitchell, A., Heymann, H., and Ebeler, S.E. (2013). Profiling the trace metal composition of wine as a function of storage temperature and packaging type. Journal of Analytical Atomic Spectrometry 28, 1288-1291
• Type: Journal Articles Status: Published Year Published: 2014 Citation: Casassa, L. F., Beaver, C. W., Mireles, M., Larsen, R. C., Hopfer, H., Harbertson, J. F., Heymann, H. (2013). The Effect of Fruit Maturity, Ethanol Concentration and Maceration Length in Washington State Merlot Wines over Two Seasons: Extraction of Phenolics and Color Components during Maceration and Sensory Implications. American Journal of Enology and Viticulture 64:437-449
• Type: Journal Articles Status: Published Year Published: 2014 Citation: King, E. S., Stoumen, M., Buscema, F., Hjelmeland, A. K., Ebeler, S. E., Heymann, H., and Boulton, R. B. (2014) Regional sensory and chemical characteristics of Malbec wines from Mendoza and California. Food Chemistry 143: 256-267
• Type: Journal Articles Status: Published Year Published: 2014 Citation: Heymann, H., Hopfer, H., and Bershaw, D. (2014). An exploration of the perception of minerality in white wines by projective mapping and descriptive analysis. Journal of Sensory Studies 29:1-13
• Type: Journal Articles Status: Published Year Published: 2014 Citation: Hopfer, H. and Heymann, H. (2014). Judging wine quality: Do we need experts, consumers or trained panelists? Food Quality and Preference 32: 221 -233
• Type: Journal Articles Status: Published Year Published: 2014 Citation: King, E.S. and Heymann, H. (2014). The effect of reduced alcohol on the sensory profiles and consumer preferences of white wine. Journal of Sensory Studies 29:33-42
• Type: Journal Articles Status: Published Year Published: 2014 Citation: Robinson, A. L., Boss, P. K., Solomon, P. S., Trengove, R. D., Heymann, H. and Ebeler, S. E. (2014). Origins of Grape and Wine Aroma: Part 1. Chemical Components and Viticultural Impacts. American Journal of Enology and Viticulture 65:1-24
• Type: Journal Articles Status: Published Year Published: 2014 Citation: Robinson, A. L., Boss, P. K., Solomon, P. S., Trengove, R. D., Heymann, H. and Ebeler, S. E. (2014). Origins of Grape and Wine Aroma. Part 2. Chemical and Sensory Analysis. American Journal of Enology and Viticulture 65:25-42
• Type: Journal Articles Status: Published Year Published: 2014 Citation: Mielby, L. H., Hopfer, H., Jensen, S., Thybo, A. K., and Heymann, H. (2014). Comparison of Descriptive Analysis, Projective Mapping and Sorting Performed on Pictures of Fruit and Vegetable Mixes. Food Quality and Preference 35:86-94
• Type: Book Chapters Status: Published Year Published: 2014 Citation: Heymann, H., King, E. S. and Hopfer, H. 2014. Classical Descriptive Analysis. In: Novel Techniques in Sensory Characterization and Consumer Profiling (Ares, G, and Varela, P, Eds.) CRC Press, New York. pp. 9-40
• Type: Book Chapters Status: Published Year Published: 2014 Citation: Hopfer, H., Ebeler, S. E. and Heymann, H. 2014. Comparison of different data analysis tools to study the effect of storage conditions on wine sensory attributes and trace metal composition. In: Food Informatics (Martinez, K., and Medina, J.) Springer, New York