Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Yeast population and community dynamics : their effect on the chemical and sensory profiles of inoculated… Neuner, Marissa 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_may_neuner_marissa.pdf [ 4.5MB ]
Metadata
JSON: 24-1.0228805.json
JSON-LD: 24-1.0228805-ld.json
RDF/XML (Pretty): 24-1.0228805-rdf.xml
RDF/JSON: 24-1.0228805-rdf.json
Turtle: 24-1.0228805-turtle.txt
N-Triples: 24-1.0228805-rdf-ntriples.txt
Original Record: 24-1.0228805-source.json
Full Text
24-1.0228805-fulltext.txt
Citation
24-1.0228805.ris

Full Text

YEAST POPULATION AND COMMUNITY DYNAMICS: THEIR EFFECT ON CHEMICAL AND SENSORY PROFILES OF INOCULATED AND SPONTANEOUS PINOT NOIR AND CHARDONNAY FERMENTATIONS AT A CANADIAN WINERY  by   Marissa Neuner   B.Sc., The University of British Columbia Okanagan, 2013     A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF   MASTER OF SCIENCE   in   THE COLLEGE OF GRADUATE STUDIES   (Biology)   THE UNIVERSITY OF BRITISH COLUMBIA   (Okanagan)   February 28, 2016     © Marissa Neuner, 2016  	 ii	Abstract The Okanagan Valley of British Columbia, Canada, has grown to include over 130 wineries, spanning 8,060 acres. As a result, competition within the local industry and has caused Okanagan winemakers to seek the production of a more sensorially unique product. Traditionally, inoculated fermentations are most commonly conducted by adding an abundance of a Saccharomyces cerevisiae yeast strain to produce consistent and specific chemical and sensory attributes in the wines; however, winemakers are choosing to showcase their unique winery yeast flora by conducting spontaneous fermentations. Spontaneous fermentations are conducted under the assumption that the fermentative yeast populations come from diverse winery resident organisms which, therefore, create different chemical and sensory profiles. Specifically, spontaneous and inoculated fermentations of Pinot noir and Chardonnay during the 2013 vintage at Quails’ Gate Estate Winery were assessed. Throughout fermentation these wines were subjected to both culture-dependent microsatellite analysis and isolate sequencing, as well as culture-independent Illumina MiSeq analysis. This determined the yeast populations conducting alcoholic fermentation and compared the capabilities of each method when studying yeast communities. Uniquely diverse yeast populations were found between spontaneous and inoculated fermentations of Pinot noir; however, yeast populations within inoculated and spontaneous Chardonnay treatments showed no differentiation. Unexpectedly, Pinot noir and Chardonnay fermentations were found to include non-Saccharomyces yeasts, specifically Hanseniaspora sp., throughout the entirety of fermentation in large relative abundance. Gas chromatography flame ionization detection (GC-FID) was conducted to determine the concentration differences of fermentation-derived chemicals associated with the identified yeast species fermenting both varietals; however, only isoamyl acetate, was found to differ between 	 iii	Pinot noir treatments. Sensory analysis revealed unique attributes were associated with the sensory profiles of inoculated or spontaneously fermented Pinot noir wines. The unique yeast communities found between inoculated and spontaneous fermentations of Pinot noir were responsible for slight chemical variation and unique sensory profiles between treatments. The similarity in yeast communities between inoculated and spontaneous fermentations of Chardonnay precluded chemical and sensory differences from occurring. Thus, to increase the chemical and sensory variation within wines and differentiate varietals from those fermented by competitors, wineries should consider producing both inoculated and spontaneous fermentations.                               	 iv	Preface  The work conducted in chapters 2 and 3 was completed in multiple stages thanks to many collaborative efforts. I was responsible for retrieving all samples from Quails’ Gate Estate Winery during the 2013 vintage across the stages of fermentation for microbial, chemical, and sensory analyses. I completed and analyzed all microbial samples in the BRAES molecular lab at the University of British Columbia’s Okanagan campus. The experimental design and chemical analyses on fermentation-derived compounds were conducted under the guidance of Dr. Sanjoy Ghosh, Dr. Wesley Zandberg and PhD. Candidate Amy Botta. The analysis of sensory data with ethics approval from the UBC Research Ethics Board (H13-03171) and Agriculture Canada was developed and collected and at the Summerland Research and Development Center (SRDC) with Kareen Stanich and Dr. Margaret Cliff.  I was responsible for the entirety of data collection, interpreting statistical analysis, and writing this thesis, under the supervision of my supervisor Dr. Daniel Durall. My supervisory committee members Dr. Margaret Cliff, Dr. Miranda Hart, and Dr. Louise Nelson have also reviewed this thesis.  All microbial microsatellite data were included in the manuscript entitled ‘Composition of Saccharomyces cerevisiae strains in spontaneous fermentations of Pinot noir and Chardonnay’ written by Scholl et al., and accepted for publication by the Australian Journal of Grape and Wine Research in November 2015. Culture dependent and independent analyses from chapter 2 will be included in the manuscript entitled ‘Comparing culture dependent and independent methods of identifying fungal communities within inoculated and spontaneous Pinot noir fermentations.’ Sensory analyses data from chapter 3 will be included in the manuscript entitled ‘Sensory attribute differentiation between inoculated and spontaneous Pinot noir fermentations.’  	 v	Table of Contents Abstract......................................................................................................................................... ii Preface........................................................................................................................................... iv Table of Contents ......................................................................................................................... v List of Tables ............................................................................................................................... ix  List of Figures................................................................................................................................ x   List of Abbreviations................................................................................................................. xiii  Acknowledgements ................................................................................................................... xiv Dedication …………………………………………………………………………………….. xvi Chapter 1 - Introduction.............................................................................................................. 1 1.1 Wine yeast diversity ………………………...…………………………………………… 1 1.2 Yeast succession during fermentation…...………………………………………………. 2 1.3 Spontaneous and inoculated methods of winemaking ...……………………………...…. 3 1.4 Identifying yeast communities during fermentation ………………………………...…... 4  1.4.1 Culture-dependent identification of yeast species and strains …………...……… 4 1.4.2 Culture-independent identification of yeast species …………………...……....... 6 1.5 Yeast-derived compounds and sensory attributes .............................................................. 8 1.6 Background of this study …………………………………………………...…….......... 11 1.7 Research objectives and hypotheses …………………………………...…………......... 12 Chapter 2 – Yeast community dynamics, composition, and diversity during inoculated and spontaneous fermentations of Pinot noir and Chardonnay at a Canadian winery.............. 14  2.1 Synopsis ………………………………………………………………………............... 14 2.2 Materials and Methods …………………………………………………………............. 16 	 vi	2.2.1 Study site, experimental design, and sampling protocol ...................................... 16 2.2.2 Culture-dependent yeast identification …………………………….................... 16   2.2.2.i    Yeast isolation and DNA extraction ...................................................... 16   2.2.2.ii   Microsatellite analysis of Saccharomyces cerevisiae strains ................. 17   2.2.2.iii  D1/D2 sequencing of non-Saccharomyces yeasts ................................. 18 2.2.3 Culture-independent yeast identification ………………………......................... 19  2.2.3.i   DNA extraction ...................................................................................... 19  2.2.3.ii   Illumina MiSeq library preparation ....................................................... 19 2.2.4 Data analysis ........................................................................................................ 21 2.2.4.i   Fermentation kinetics and diversity ....................................................... 21 2.2.4.ii   Illumina sequence analysis .................................................................... 21 2.2.4.iii  Community structure ............................................................................. 22 2.3 Results .............................................................................................................................. 23 2.3.1 Fermentation of inoculated and spontaneous Pinot noir and  Chardonnay .......................................................................................................... 23 2.3.2 Yeast succession throughout Pinot noir and Chardonnay fermentations ............. 24 2.3.3 Yeast strain and species identities found by culture dependent analyses ............ 27 2.3.4 Culture dependent and independent species richness and community structure.. 30 2.3.5 Diversity of yeast species ..................................................................................... 34 2.4 Discussion ........................................................................................................................ 36  2.4.1 Yeast dynamics during inoculated and spontaneous fermentation ...................... 36  2.4.2 Non-Saccharomyces and S. cerevisiae succession during fermentation ............. 37  2.4.3 Yeast species and commercial strain identification ............................................. 38 	 vii	 2.4.4 Species richness, community structure, and diversity as per culture dependent and independent analyses ........................................................................................... 40 Chapter 3 - Chemical and sensory differentiation of spontaneous and inoculated Pinot noir and Chardonnay fermentations at a Canadian winery .......................................................... 44 3.1 Synopsis ………………………………………………………………………………... 44 3.2 Materials and methods ..................................................................................................... 46  3.2.1 Sampling protocol ................................................................................................ 46 3.2.2 Chemical analysis of wines .................................................................................. 46  3.2.2.i  Enological analysis of wines ................................................................... 46  3.2.2.ii  GC-FID analysis and standard curve preparation ................................... 48 3.2.3 Sensory analysis of wines .................................................................................... 49  3.2.3.i  Racking and bottling of wines ................................................................ 49  3.2.3.ii  Sensory standard preparation .................................................................. 50  3.2.3.iii Aroma, flavor by mouth, and colour analysis ........................................ 52 3.2.4 Data analysis ………………………………………...…………………............. 53  3.2.4.i  Standard concentration curves ................................................................ 53  3.2.4.ii  GC-FID analysis ..................................................................................... 53  3.2.4.iii Sensory analysis ..................................................................................... 53 3.3 Results …………………………………………………………………………….......... 55 3.3.1  Enological and GC-FID analyses of wines .......................................................... 55 3.3.2 PCA and cluster analyses of yeast-derived compounds ...................................... 57 3.3.3 Aroma, flavour by mouth, and colour attribute intensities ................................. 59 3.3.4 Sensory attribute differentiation of inoculated and spontaneous fermentations .. 61 	 viii	3.4 Discussion ………………………………………………………………………............ 65 3.4.1 Chemical differentiation by yeast communities of inoculated and spontaneous fermentations .................................................................................................................... 65 3.4.2 Sensory differentiation by yeast communities of inoculated and spontaneous fermentations .................................................................................................................... 68 Chapter 4 – Conclusion ……………………………………………………...……….............. 70 4.1 Summary ...................……………………………………………...……………............ 70 4.2 Novelty of the research …………………………………………………………............ 73 4.3 Management implications …………………………………………………...…............. 73 4.4  Suggestions for further research ………………………………………..………............ 74 References ……………………………………………………………………………............... 77 Appendices ……………………………………………………………………………….......... 89 Appendix A: Fermentation parameters …………………………………………............ 89 Appendix B: Culture dependent and independently identified yeast species and strain relative abundances ..................................................................................... 93 Appendix C: Three-way ANOVAs for Pinot noir and Chardonnay sensory evaluations 97 	 ix	List of Tables  Table 1.1 Yeast-derived compounds, known aroma detection thresholds, and sensory       descriptive terms ........................................................................................................... 9 Table 3.1 Sensory standards created for Pinot noir and Chardonnay aroma and flavour by    mouth assessments ...................................................................................................... 51 Table 3.2 Mean concentrations and standard error (SE) of primary yeast-derived compounds within inoculated and spontaneous fermentations of Pinot noir and Chardonnay  as detected by GC-FID analysis (n=6) ...................................................................... 56 Table 3.3 Mean attribute intensity scores and standard error (SE) for inoculated and         spontaneous fermentations of Pinot noir and Chardonnay. Attributes were  assessed in duplicate and rated on a 0-100 intensity scale by a panel of  8 experts (n=48).......................................................................................................... 60   	 x	List of Figures Figure 2.1 Fermentation kinetics of inoculated (I) and spontaneous (S) fermentations of  A) Pinot noir and B) Chardonnay (n=3). Colony forming units per milliliter (CFU/mL) and °Brix was measured and compared at each stage of  fermentation .............................................................................................................. 24 Figure 2.2 Relative abundances of Saccharomyces cerevisiae and non-Saccharomyces  isolates identified using microsatellite and sequencing analyses within  A) inoculated and B) spontaneous Pinot noir fermentations, and C) inoculated  and D) spontaneous Chardonnay fermentations (n=3) ............................................. 26 Figure 2.3 Average strain and species abundance as determined by microsatellite and sequencing analyses during all fermentation stages of A) Pinot noir inoculated  with Lalvin RC212 and B) spontaneous Pinot noir, and C) Chardonnay  inoculated with Zymaflore VL2 and D) spontaneous Chardonnay (n=3).  All yeast strains used as commercial inocula concurrently or previously by the winery are shown in bold .......................................................................................... 29 Figure 2.4 Expected and observed relative abundances of the yeast species H. uvarum, T. delbruekii, S. cerevisiae, and S. uvarum present within the mock community ......... 30 Figure 2.5 Relative abundances of dominant species (greater than 1% relative abundance) present within all stages of A) inoculated Pinot noir, B) spontaneous Pinot noir,  C) inoculated Chardonnay, and D) spontaneous Chardonnay as determined by  culture dependent (i) and independent (ii) methods .................................................. 32   	 xi	Figure 2.6 Figure 2.6. NMS ordinations of total yeast species composition based on Bray- Curtis dissimilarity matrices. Ordinations are reported for inoculated and spontaneous fermentations of A) Pinot noir and B) Chardonnay for both culture dependent (i) and independent (ii) methods with significant treatment differences occurring between Pinot noir treatments .................................................................. 33 Figure 2.7 Simpson’s Index of Diversity for all stages of fermentation between inoculated  and spontaneous treatments of A) Pinot noir and B) Chardonnay as determined  by culture-dependent (i) and culture-independent (ii) microbial identification methods. Significant differences in diversity were determined by a two-way  ANOVA and indicated by * ...................................................................................... 35 Figure 3.1 Average peak areas of all standard compounds at 50, 100, and 150 mg/L and  the resulting standard curve equations (n=8) ............................................................ 49 Figure 3.2 A) Titratable acidity, B) volatile acidity, C) residual sugar (°Brix), D) pH, and E) alcohol content of late-stage inoculated and spontaneous Pinot noir and  Chardonnay fermentations. Significant differences (p ≤ 0.05) in chemical  attributes are indicated by * ...................................................................................... 52 Figure 3.3 Principal component and hierarchical cluster analyses using the Ward’s method  of yeast-derived compound concentrations within inoculated (I) and spontaneous  (S) fermentations of A) Pinot noir and B) Chardonnay (n=12). Tanks (T1-3) and replicates (R1-2) are denoted for all samples in PCA analyses, with the average  of the two replicate samples per tank being utilized for cluster analysis .................. 55 	 xii	Figure 3.4 Radar plots illustrating differences between mean intensity scores for all  attributes associated with inoculated (black) and spontaneous (blue)  fermentations of A) Pinot noir and B) Chardonnay. Significant differences  (p ≤ 0.05) are illustrated by * .................................................................................... 60 Figure 3.5 Principal component and hierarchical cluster analysis using the Ward’s method  of most significantly different sensory attribute intensities associated with  sensory profiles of inoculated (I) and spontaneous (S) fermentations of  A) Pinot noir and B) Chardonnay (n=12). Tanks (T1-3) and replicates (R1-2)  are denoted for all samples in PCA analyses, with the average of the two  replicate samples per tank being utilized for cluster analysis ................................... 62    	 xiii	 List of Abbreviations  ADY  Active dry yeast  ANOVA Analysis of variance BLAST Basic local alignment search tool BRAES Biodiversity resilience and ecosystem services BSA  Bovine serum albumin DNA  Deoxyribonucleic acid dNTPs  Deoxynucleoside triphosphate FADSS Fragment analysis and DNA sequencing services FLASh Fast length adjustment of short reads GC-FID Gas chromatography flame ionization detection ITS  Internal transcribed spacer NMS  Nonmetric multidimensional scaling PCA  Principal Component Analysis PCR  Polymerase chain reaction QGEW Quails’ Gate Estate Winery TA  Titratable acidity UBCO  University of British Columbia Okanagan VA  Volatile acidity YEPD  Yeast extract peptone dextrose     	 xiv	Acknowledgements I wish to share my sincerest gratitude to all those who have supported me through the completion of this work. Specifically, I wish to thank my supervisor Dr. Daniel Durall for allowing me to be involved in this project, his unconditional guidance and support, and challenging me to pursue my absolute best. I would also like to thank my supervisory committee members, Dr. Margaret Cliff, Dr. Miranda Hart, and Dr. Louise Nelson for providing me with exceptional guidance, extensive resources and advice as I conducted my data collection and analysis. I also wish to thank my generous funding sources, the Natural Sciences and Engineering Research Council, Quails’ Gate Estate Winery, and the BC Wine Grape Council. I would also like to thank all of my gracious collaborators for allowing me to conduct research in their labs, as well as for teaching me how to approach my project from a variety of angles, opening my eyes to a scope of scientific disciplines I would never have imagined being a part of. To Drs. Sanjoy Ghosh, Wesley Zandberg, Sandra Mecklenburg, and James Bailey, and Amy Botta (PhD. Candidate), thank-you for all your guidance during the development of my GC-FID analysis. To Cameron Egan (PhD. Candidate) and IT specialist Wade Klaven, thank-you for such unwavering support during the development and analysis of my next-generation microbial work. Lastly, to everyone who participated in my sensory panel, and specifically to Kareen Stanich, thank-you for contributing your time and efforts into generating sensory profiles of my wines.  I wholeheartedly wish to extend my gratitude to the staff of Quails’ Gate Estate Winery, for allowing me into their cellar and laboratory during the 2013 vintage, while providing me with endless resources and support as I conducted this research. I thank winemaker Nikki Callaway for trusting me with her product and providing me with laboratory access to conduct various 	 xv	analyses on site. Also, to David Ledderhof, who ensured the experimental design was met, and safekeeping of research vessels during vintage.  Lastly, and most importantly, I wish to thank every member of the Durall lab and UBCO biology graduate student family I had the privilege of sharing my time with. To Chrystal Scholl, Morgan Stone, Sydney Morgan, Mansak Tantikachornkiat, and Stacey Sakakibara, thank-you for always helping me to brainstorm, for the sanity you provided during the late nights, and the unconditional comradery.   	 xvi	Dedication I wish to formally dedicate this thesis to my mentors, family and friends, without whom I would never have found the optimism, strength, and motivation to persevere. 	 1	Chapter 1 – Introduction 1.1 Wine yeast diversity Many studies have been conducted to understand the commonly found vineyard and winery resident yeast species. Primarily, non-Saccharomyces species dominate the vineyard flora with the common genera being Hanseniaspora, Candida, Metschnikowia, Rhodotorula, Cryptococcus, and Pichia (Clavijo et al. 2010; Barata et al. 2012). However, Saccharomyces cerevisiae strains are also present within vineyards, but they are usually not high in abundance (Barnett et al. 1972; Le Jeune et al. 2006; Mercado et al. 2007). Studies have been conducted to specifically determine the yeast microbiomes within wineries such that species persistence in various winery environments can be better understood. Typically, these yeasts are commercial Saccharomyces cerevisiae inocula used in previous vintages that have settled on winery surfaces creating “winery resident” yeast populations (Martini 2003). Most commonly, yeast can persist on tank surfaces, barrels and on crushing equipment where nutrients are plentiful despite steam cleaning between vintages (Mercado et al. 2007; Bokulich et al. 2013b). Due to the porous nature of barrels it is difficult to sterilize them between vintages, thus spontaneous fermentation (see section 1.3) relying on persistent yeasts is quite common in barrel-fermented varietals (Blanco et al. 2011). It has also been investigated whether winery age influences the resident microflora of a winery with the literature suggesting that older cellars have a more diverse microflora than do younger cellars (Santamaria et al. 2005). Furthermore, it has been suggested that the yeast diversity of the winery may influence the dominant strains fermenting the wine must (Lange et al. 2014). A rising trend in winemaking is the isolation and culture of both non-Saccharomyces species and Saccharomyces yeast strains as commercial inocula. Non-Saccharomyces species 	 2	have been shown to influence fermentations much more than previously thought; therefore, their addition as commercial yeast inocula has increased substantially (Ciani et al. 2010; Comitini et al. 2011; Zott et al. 2011). As they are not capable of completing alcoholic fermentation of wine must, co-inoculation of non-Saccharomyces species with a specific S. cerevisiae strain has become more prevalent. These organisms are chosen so that the sensory attributes produced by the yeast inoculum best suits the must characteristics of the grape varietal (Esteve-Zarzoso et al. 1998; Comitini et al. 2011; Suarez-Lepe and Morata 2012). More commonly, inoculation with Saccharomyces cerevisiae strains alone has become the standard for winemaking. There are many companies that provide hundreds of commercially produced pure cultures of S. cerevisiae active dry yeast (ADY) inocula. These yeasts have been created based on their metabolic and sensory output in wine (Mortimer and Polsinelli 1999; Legras et al. 2007). As a result, commercial yeast-producers select for strain variants that produce different sensory and chemical characteristics under specific conditions. These behavioral variants are then cultured and marketed as a new strain, which is better suited to the specific must conditions (Querol et al. 2003). As winemakers seek to obtain desired chemical and sensory characteristics of grape must, they can identify an ideal yeast inoculum that will produce chemical and sensory attributes to best complement each varietal.  1.2 Yeast succession during fermentation  As alcoholic fermentation progresses, the environment of the grape must changes dramatically. Initially, it is a sugar and nutrient-rich environment, with low levels of toxic compounds, that is capable of sustaining many different yeast species; however, as alcoholic fermentation begins, the alcohol-intolerant non-Saccharomyces species tend to die due to alcohol toxicity (Dimaro et al. 2007; Hall et al. 2011). Saccharomyces cerevisiae is very alcohol tolerant 	 3	and is capable of surviving in musts of up to 16-18% alcohol. Because S. cerevisiae competes effectively with other yeast species at high alcohol levels, it can consume the remaining fermentable sugars, which increases cell number to their highest level (Guzzon et al. 2011). As resources are depleted, non-Saccharomyces yeasts usually disappear in the middle and end stages of fermentation. Recently, however, this dynamic succession of yeasts has been partially challenged in that certain non-Saccharomyces species may be more alcohol tolerant than once thought and may have a larger influence on the wine throughout fermentation (Torija et al. 2001). Species such as Pichia, Rhodotorula and Metschnikowia have been shown to persist and remain metabolically active at the end of fermentation, despite alcoholic fermentation reaching completion (Diaz et al. 2013).  1.3 Spontaneous and inoculated methods of winemaking  Traditionally, winemaking originated as an entirely spontaneous process, relying on the yeasts found naturally occurring on grape and winemaking surfaces to conduct fermentation (Mortimer and Polsinelli 1999; Legras et al. 2007; Cordero-Bueso et al. 2011). These spontaneous fermentations simply involve crushing the grapes and allowing fermentation to occur without directed or guided inoculation, which tends to result in a longer fermentation time but can make a unique and complex wine that reflects the vineyard and winery conditions of that single year (Mercado et al. 2007). Unfortunately, spontaneously produced wines are more prone to the establishment of spoilage organism communities because there is an abundance of nutrient resources in the must and limited microbial competition. Therefore, in modern winemaking, the use of these fermentations is limited because in some regions it has been shown that additional monitoring is necessary to maintain a healthy fermentation and this extra effort may not be feasible (Loureiro and Malfeito-Ferreira 2003). Alternatively, as winemaking has been 	 4	conducted since antiquity, long before the identification of the organism responsible for fermentation was discovered, inoculation is a relatively new but common practice for current winemakers. This process relies on the purposeful addition of a large amount of active dry yeasts to grape must with the specific intention of imparting characteristics associated with that yeast strain into the resulting wine (Barrajon et al. 2010). The fermentative yeast Saccharomyces cerevisiae has been studied in depth as it pertains to wine fermentations, allowing for the generation of select strains that are better suited to ferment certain grape varietals, impart specific attributes and survive adverse cellar conditions (Querol et al. 2003). While this should allow for the reliable and rapid production of wines with little spoilage risk, such an abundance of the same yeast population has been thought to create wines that are ‘one-note’ and may limit the range of chemical and sensory attributes experienced when drinking the wine (Saberi et al, 2014). Recently, as competition in the Okanagan wine industry is increasing, winemakers must carefully decide which methods best suit their production needs, prompting the need for studying both inoculated and spontaneous yeast communities present within fermentations of many grape varietals.  1.4 Identifying yeast communities during fermentation 1.4.1 Culture-dependent identification of yeast species and strains Many methods of identifying microorganisms have been used in the past; however, some of the most reliable identification techniques first require that live organisms be cultured on appropriate media prior to identification. While this ensures that organisms are alive and potentially metabolically active within a sample, determining the microbial community of a wine sample in this manner is subject to many biases. Culture-dependent techniques take significant time and resources, therefore lower numbers of isolates can be processed and identified, which 	 5	may improperly represent the microbial community. Also, plating limits one from detecting non-culturable and slower-growing species that may actually be abundant. However, culture-dependent identification has allowed for much more specific identification of yeast isolates, which has proven beneficial to the wine industry. Because fermentation is conducted by strains of the same species of yeast, Saccharomyces cerevisiae, culture-dependent means of identification, such as microsatellite analysis, are presently the only way to determine the assemblage of S. cerevisiae strains (Richards et al. 2009)  Microsatellites are short DNA sequence repeats that are highly variable between alleles of different organisms that can act as genetic markers to distinguish between strains of a single species. They are particularly useful for strain identification due to the ease at which they can be amplified by using traditional polymerase chain reaction (PCR) techniques (Goldstein and Schlotterer 1999). The production of databases that contain allele combinations from different hyper-variable loci of different strains has allowed researchers, studying environmental samples, to distinguish between different commercial and non-commercial S. cerevisiae strains (Legras et al. 2005; Richards et al. 2009; Hall et al. 2011).  Microsatellites allow for the creation of a genetic fingerprint unique to a specific S. cerevisiae strain so that future studies involving these strains can be compared (Hannequin et al. 2001). Most of the other identification methods currently available are only able to distinguish different species present in a population and are not sensitive enough to differentiate individual strains of the same species. This is particularly problematic for studying yeast populations in wine, because there are hundreds of commercially available S. cerevisiae strains. Only by determining the microsatellite fingerprint unique to each individual strain, can these differing strains of a single species be detected. To increase efficiency of future yeast species research, Richards et al. (2009) developed a system for 	 6	genotyping Saccharomyces cerevisiae strains based on ten microsatellite loci that have unique alleles specific to single commercial strains. This database outlines microsatellite fingerprints for many commercial and non-commercial strains of yeasts; however, there are many more commercial strains used across the world that could be added to the database. In addition to the Richards et al. database (2009), a local Okanagan commercial yeast database has been produced, which is based on multiple microsatellite loci (Hall et al. 2011), and has been recently updated (Scholl et al. 2016). Identifying non-Saccharomyces spp. involves extracting the fungal DNA and amplifying a specific region to be submitted for sequencing. This sequence is compared to sequences previously submitted to the Basic Local Alignment Search Tool (BLAST) database such that the identity of the sample is obtained (Schoch et al. 2012). The D1/D2 domain of the large ribosomal DNA subunit is highly polymorphic among yeast species and it can serve as an identifying sequence (Lachance et al. 2003). By amplifying this region, non-Saccharomyces yeast species can be distinguished, allowing for the identification of the many species present during the cold-soak stage of fermentation and beyond. Specifically, cold soaking involves the crushing of grapes which soak with grape solids, stems and/or skins at low temperatures which prevent fermentation such that grape compounds can be extracted, and particulates settle out; therefore, this environment is rich in yeast species that may establish populations throughout the subsequent fermentation stages (Li et al. 2012).  1.4.2 Culture-independent identification of yeast species While the aforementioned identification methods are extremely useful, they are not completely accurate in the representation of species diversity within a sample. This is because microsatellite and sequencing analysis relies on extracting DNA from individual organisms, 	 7	whereas next-generation techniques, using a culture-independent approach, allow for the creation of a pooled library of species DNA. These techniques have accelerated sequencing time, decreased costs and increased the number of DNA sequences that can be determined in a single DNA library. This substantially benefits many areas of environmental and industrial research that require studying communities consisting of many different organisms (Mardis 2008; Schuster 2008; San Miguel 2011).  Recently, next generation sequencing using the Illumina Genome Analyzer (Illumina) has increased in popularity among scientists because it has allowed for the sequencing of entire microbial communities within a single sample. Illumina involves the production of a genomic DNA library by extracting and amplifying the desired DNA directly from the sample. Illumina-specific adapter sequences bind the amplified DNA to the inside channels of the sequencing flow cells. Polymerase and fluorescently labeled nucleotides are added to the flow channels initiating a bridge amplification of the DNA strands present in the library. Following the incorporation of a single nucleotide, the unused nucleotides and polymerase are washed away. Chemicals that cleave the fluorescent markers are added, identifying the incorporated nucleotide of each DNA fragment. This cycle repeats until approximately a million copies of the original fragment are bound to the flow cell channel that consist of 100-150 base-read lengths (Mardis 2008). Recently, there have been many advances in the high throughput and next generation approaches to identify microbial populations in foods and beverages, specifically targeting fungi and lactic acid bacteria (Bokulich and Mills 2012, 2013; Bokulich et al. 2013b). By identifying a novel internal transcribed spacer (ITS) primer set that distinguishes fungal species, the individual yeast community members can be determined in wine fermentations (Bokulich and Mills 2013). 	 8	Nevertheless, to ensure accurate reporting of sequence identification and its relative abundance, the construction of a mock reference community is crucial.  1.5 Yeast-derived compounds and sensory attributes  Saccharomyces cerevisiae is primarily responsible for converting glucose to ethanol, and for allowing fermentation to reach completion. It is the main yeast species capable of enduring alcohol toxicity and consuming all fermentable sugars, while producing minimal negative and many positive secondary metabolites (Capece et al. 2010). In spontaneous fermentations, the number of secondary metabolic chemicals produced in wine has been shown to positively correlate with yeast diversity due to the variety of expressed metabolic pathways (Camarasa et al. 2011). The production of chemicals in wine by S. cerevisiae has been extensively detailed because this organism has been shown to produce six different classes of volatile chemicals that significantly influence the chemical attributes of wines through the metabolism of inorganic and organic molecules: alcohols, acids, esters, carbonyls, sulphur-containing volatiles and thiols (Ugliano and Henschke 2009; Saerens et al. 2010; Sumby et al. 2010; Saberi et al 2012). However, research suggests that non-Saccharomyces yeasts can significantly influence the chemical profile of wines. Despite their assumed susceptibility to alcohol toxicity, they have been shown to persist during later stages of fermentation (Ciani and Maccarelli 1998; Plata et al. 2003; Romano et al. 2003). Prominent vineyard non-Saccharomyces yeasts, Hanseniaspora, Candida, Pichia and Metschnikowia, have only recently been studied with respect to their metabolic capabilities and chemical contribution to the final wine product. Hanseniaspora, the most abundant vineyard yeast and most common inhabitant of cold-soak, has been shown to contribute towards favourable levels of higher alcohols and volatile esters, such as isoamyl acetate during 	 9	spontaneous fermentations (Ciani and Maccarelli 1998; Romano et al. 2003; Ciani et al. 2010; Moreira et al. 2011). Candida species have been shown to slightly enhance glycerol and alcohol content, but during cold-soak and in the absence of ethanol, they are capable of producing 100 times the ethyl acetate produced by S. cerevisiae (Plata et al. 2003). Pichia species have been shown to positively enhance isoamyl acetate, but most predominantly, produce ethyl acetate (Plata et al. 2003; Ciani et al. 2010). Ultimately, the production of certain higher alcohols, esters and acetates that may be present in concentrations capable of influencing the sensory profiles of wines is reliant on the diversity of yeasts present during fermentation (Table 1.1). Thus, it is imperative to consider the entire non-Saccharomyces and S. cerevisiae yeast community structure when comparing it to the resulting wine chemistry (Swiegers et al. 2005, Ugliano and Henschke 2009, Styger et al. 2011).  Table 1.1. Yeast-derived compounds, known aroma detection thresholds, and sensory descriptive terms Compound Aroma threshold (mg/L) Descriptive terms Acetaldehyde ~0.5a Pungent, bruised applea Propanol ~500b Harshb Ethyl acetate ~12a Pineapplea, tropicalc Isobutanol  ~40a Solventa, bitterd Amyl alcohol ~30e Sweet, fruity Isoamyl alcohol  ~30a Malt, burnta Isoamyl acetate  ~0.03a Bananaa a Francis and Newton. 2005 b Swiegers et al. 2005 c Campo et al. 2005 d Cullere et al. 2004 e Molina et al. 2007  As previously stated, both Saccharomyces and non-Saccharomyces yeasts are capable of imparting many compounds into wines due to the many metabolic pathways they employ during 	 10	fermentations; however, in the Okanagan region it is relatively unknown how the sensory attributes of spontaneous and inoculated fermentations of the same grape varietal differ, and if these differences can be attributed to specific chemical contributions of the yeast community within each wine. Studies following the aromatic and flavour contribution of S. cerevisiae strains to wines have generated a vast array of sensory descriptors pertaining to volatile aromatic compounds (Mateos et al. 2006; Vilanova and Sieiro 2006; Bisson and Karpel 2010). Alternatively, a surge in research has been conducted on fermentations with non-Saccharomyces yeasts found to persist during fermentation, specifically Hanseniaspora, Pichia, and Brettanomyces ssp., which have been shown to produce many different volatile aromatic and flavour compounds that influence the sensory profile of wines as well (Romano et al. 2003; Moreira et al. 2011).  Evidence suggests that the aroma and flavour profiles of spontaneously produced wines can be more unique, diverse, and complex than inoculated wines (Varela et al. 2009; Saberi et al. 2014). This ‘complexity’ is likely due to a high diversity of compounds produced by a diverse yeast community. Inoculated wines are thought to be lower in yeast diversity and may have sensory profiles consisting of dominant attributes due to a low diversity of compounds produced by one or a few dominant yeasts. However, for the Okanagan region, the sensory differences of spontaneous and inoculated fermentations are not yet known. It is crucial to understand the yeast communities among these fermentations such that resulting sensory profiles can be attributed to the different communities responsible. Knowing this information may help to determine whether wine production could benefit from altering production to a more spontaneous fermentation approach as opposed to the present inoculated approach used by most winemakers.  	 11	1.6 Background of this study  The cultural importance of wine, both historically and currently, is unparalleled by almost any other food or beverage. This fundamental connection of wine with the human population is evidenced by the historical spread of wine across the globe (Legras et al. 2007). By understanding the influence of wine-yeast species and strains on the chemistry and sensory profiles of different fermentation styles, a more unique, enjoyable and ultimately profitable product can be enjoyed by consumers. In the Okanagan specifically, it is relatively unknown how yeast populations from the vineyard and winery interact to influence the chemistry and sensory profiles of our local wines. As a result, it is unclear whether we produce distinct wines that reflect and differentiate our wine region from others and ultimately whether our current method of inoculated fermentation is an ideal approach to produce premium wines. Previous research within local wineries has indicated that inoculated yeasts dominate spontaneous fermentations, suggesting inoculation may not be necessary following the establishment of winery yeast genera of past inoculations (Hall et al. 2011; Lange et al. 2014). Simultaneously, little is known of the specific ways in which inoculated and spontaneous fermentations differ among non-Saccharomyces species present and their chemical and sensory input. Only by understanding the effect all yeast populations have on the final fermented product, will we be able to modify and properly manage alcoholic fermentations. The information gained from this research will ultimately insure the production of successful vintages in the future.         	 12	1.7 Research objectives and hypotheses   Objective 1: Use culture-dependent and culture-independent identification methods to determine yeast diversity and composition within inoculated and spontaneous fermentations of Pinot noir and Chardonnay wines at a single winery. • Hypothesis 1-1- Inoculated as compared with spontaneous fermentation will have higher CFU/mL at the early, mid, and end stages due to the purposeful addition of the inoculum  • Hypothesis 1-2- Non-Saccharomyces species will be the dominant yeasts in cold-soak, but S. cerevisiae will dominate in the early, mid, and end stages of fermentation • Hypothesis 1-3- Spontaneous fermentation will be largely driven by S. cerevisiae strains previously or concurrently used as commercial inocula by the winery that enter the must • Hypothesis 1-4- Culture dependent and independent approaches will report similar results in terms of dominant species richness, but relative abundance values and overall community structure will differ • Hypothesis 1-5- Spontaneous fermentation as compared with inoculated will have a higher diversity of yeasts for both culture dependent and independent approaches Objective 2: To determine how the chemical and sensory attributes differ between inoculated and spontaneous fermentations of Pinot noir (as they were fermented by significantly different yeast communities), and of Chardonnay (where the yeast communities were similar for the two types of fermentations). • Hypothesis 2-1- Volatile acidity, residual sugar, and alcohol content will differ between inoculated and spontaneous fermentations of Pinot noir because these are greatly influenced by the different yeast communities fermenting the must, while pH and titratable acidity will remain similar due to the greater influence of the must on these 	 13	characteristics. No differences will be seen in Chardonnay because no community differences were present between inoculated and spontaneous fermentations. • Hypothesis 2-2- Yeast-derived compound concentrations will differ between inoculated and spontaneous fermentations of Pinot noir because they were fermented by different yeast communities, and show different associations with each fermentation treatment. There will be no difference between Chardonnay treatments because community differences were not detected. • Hypothesis 2-3- Inoculated Pinot noir wines will have more significantly higher sensory attribute intensities than spontaneous wines due to lower yeast diversity. There will be no difference in attribute intensities between Chardonnay wines because no differences in diversity occurred. • Hypothesis 2-4- Inoculated and spontaneous fermentations of Pinot noir will be differentiated by certain sensory attributes associated specifically with each treatment due to their different yeast communities. There will be no sensory differentiation between Chardonnay treatments.  	 14	Chapter 2 – Diversity and community composition dynamics of yeasts during inoculated and spontaneous fermentations of Pinot noir and Chardonnay at a Canadian winery.  2.1 Synopsis The Okanagan valley wine region has been growing substantially in recent years, creating a competitive market in which winemakers must seek out ways to differentiate themselves from producers of the same varietals. One way winemakers can do this is to modify or manipulate the yeast community producing the wine. However, there is limited knowledge pertaining to species and strain dynamics of yeasts and how they affect the aroma and flavour of the final wine produced. Spontaneous and inoculated methods of winemaking have both positive and negative aspects that influence the wines produced. Inoculated fermentations are thought to be dominated by commercial active dry yeast (ADY), therefore, they tend to be reproducible, allowing a consistent product to leave the cellar annually (Rankine and Lloyd 1963; Santamaria et al. 2005). Furthermore, because of its dominance, the ADY tends to out-compete spoilage organisms and prevent them from flourishing. Thus, product loss tends to be minimal under inoculated conditions (Loureiro and Malfeito-Ferreira 2003). On the other hand, spontaneous fermentations, while they may be more prone to spoilage due to the lack of competition from a Saccharomyces cerevisiae inoculum, are thought to produce wines that are more complex and unique in their aroma and flavour by mouth attributes. As compared to inoculated fermentations, the abundance of sensory attributes in spontaneously fermented wine is likely due to the diverse yeast community allowed to populate the must. Simultaneously, the complexity may arise from commercial yeast choices of a winemaker from year to year, strains native to the vineyard and winery environment to which the grape must is exposed, and the grape varietal and vineyard location (Le Jeune et al. 2006; Hall et al. 2011; Blanco et al. 2013). Traditionally, it was thought 	 15	that non-Saccharomyces yeasts would be dominant prior to inoculation (by ADY or spontaneously) and they would be replaced completely by the alcohol tolerant Saccharomyces cerevisiae by the end of fermentation. Nonetheless, the typical succession of yeast populations thought to be indicative of all winery fermentations may not actually be occurring in all fermentations. Recently, it was shown that certain non-Saccharomyces species have been increasing in abundance during the later, more alcoholic stages of fermentation (Ciani and Maccarelli 1998; Comitini et al. 2011; Moreira et al. 2011). Therefore, it is imperative to understand the dynamics of yeast communities within both spontaneous and inoculated wines fermented in tank and barrel, such that yeast prevalence during fermentation can be updated to reflect their current capabilities. Currently, the most widely used method of identifying wine yeasts is culture-dependent, meaning grape must directly sampled from the fermentations is plated onto culture medium, and yeast isolates are identified to species and strain level; however, this analysis is limiting as it may underestimate the richness and abundance of species and strains by preventing viable but not culturable, slow growing, and/or less competitive yeasts from being isolated onto culture medium. Consequently, there is a movement towards identifying the entire yeast community of a sample through next-generation sequencing on the Illumina MiSeq® platform, which may not determine if an organism is live or dead at the time of sampling, but will identify all species present. The objective of this study was to compare non-Saccharomyces and S. cerevisiae dynamics between inoculated and spontaneous fermentations at an Okanagan winery using culture dependent and independent identification methods.  	 16	2.2 Materials and methods 2.2.1  Study site, experimental design, and sampling protocol  This study was conducted during the 2013 vintage at Quails’ Gate Estate Winery, in West Kelowna, British Columbia, Canada. Grapes for each treatment were harvested from neighbouring blocks in the same Pinot noir vineyards (Appendix A Tables A1-4). Six tanks ranging between 3000-5300 L were filled with grape must, three of which were inoculated with S. cerevisiae inoculum strain Lalvin® RC 212 and three were left to spontaneously ferment. Simultaneously, six 225 L French oak barrels were filled with Chardonnay grape must, three of which were inoculated with S. cerevisiae inoculum Zymaflore VL2 and three were left to spontaneously ferment. Each replicate tank at four distinct stages of fermentation, as defined by Brix levels (cold soak stage= ≥ 21°, early stage=15-20°, mid stage=3-15°, late stage= ≤2° brix). were sampled. At each sampling stage, following pumpover for tanks and a thorough stirring for barrels, two 50 mL sub-samples were taken from 1.5 m depth (per tank), or from 30 cm in depth (per barrel). Samples were placed on ice and taken to UBC Okanagan where yeast cell count was determined using a hemacytometer as per Iland at al. (2007). Appropriate dilutions from the first sub-sample were spread onto plates containing yeast extract peptone dextrose (YEPD) medium, and incubated at 28°C for 24-48 hours to produce a plate of 30-300 colonies for culture-dependent analysis. The remaining sub-sample was aliquoted into 25 2 mL centrifuge tubes for culture-independent analysis (see section 2.2.3) and stored at -80°C.  2.2.2  Culture dependent yeast identification 2.2.2.i   Yeast isolation and DNA extraction Following incubation, 24 yeast colonies were randomly selected from plates containing 30-300 colonies and streak plated. Thus, 24 isolates per tank or barrel at each sampling stage 	 17	were identified (24 isolates per stage x 4 stages of fermentation x 12 fermentations = 1,152 identified isolates). Single isolates from the streak plates were placed in 96-well PCR plates and submerged in 50 µL of DNA-grade sterile water (Fisher BioReagents, Pittsburgh, USA). Following vortexing, each plate was incubated at 95°C in an Applied Biosystems® Veriti® 96-Well Fast Thermal Cycler (Foster City, USA) for 15 min to lyse cells and release DNA.  Samples were stored at -20°C until further analysis. 2.2.2.ii  Microsatellite analysis of Saccharomyces cerevisiae strains Saccharomyces cerevisiae strain isolates were identified by amplifying seven hyper-variable microsatellite loci (C4, C8, C3, C11, SCY, YPL, and YML) with forward primers labeled with fluorescent dye at the 5’ end (Field and Wills 1998; Perez et al. 2001; Legras et al. 2005). This 7-plex PCR contained the following reaction conditions: 1 X colorless GoTaq® reaction buffer (Promega©, Madison, USA), 9.0 nmol of each dNTP, 0.9 pmol of C3 forward and reverse primer, 12.7 pmol of C4 forward and reverse primer, 4.3 pmol of C8 forward and reverse primer, 2.2 pmol of C11 forward and reverse primer, 1.2 pmol of SCY forward and reverse primer, 7.8 pmol of YML forward and reverse primer, 1.5 pmol of YPL forward and reverse primer, 10.0 µg bovine serum albumin (BSA), 35 nmol MgCl2, 1.25 U of GoTaq® DNA Polymerase (Promega©, Madison, USA), and 5 ng genomic yeast DNA. The final volume was adjusted to 15.0 µL using DNA-grade sterile water (Fisher BioReagents, Pittsburgh, USA). Amplification was performed using an Applied Biosystems® Veriti® 96-Well Fast Thermal Cycler (Foster City, USA) with the following parameters: 94°C for 3 min (1 cycle), 94°C for 30 s, 55°C for 35 s, 72°C for 45 s (36 cycles) and 72°C for 10 min (1 cycle). Fragment size of each locus was determined using a 3130xl DNA sequencer (Applied Biosystems, Foster City, CA, USA), which resulted in a DNA fingerprint for each isolate. Gene Mapper 4.0 software (Applied 	 18	Biosystems, Foster City, CA, USA) was used to determine the strain-type for each fingerprint. To obtain commercial strain identity, genotypes were compared with a previously published database (Richards et al. 2009) and with our own database. The estimated probability that two unrelated strains could have identical multi-locus genotypes by chance was 3.7x10-7. This value was calculated by using GenA1Ex software 6.1 (www.anu.edu.au/BoZo/GenA1Ex/).  2.2.2.iii D1/D2 sequencing of non-Saccharomyces yeasts All cold soak isolates and those that did not provide a microsatellite fingerprint across the remaining stages of fermentation were considered to be non-Saccharomyces species. The species of yeast was determined by DNA amplification of the D1/D2 region of the large ribosomal RNA subunit with NL primers. The final volume for all NL reactions was 12.5 µL and included the following reagents: 6.17 µL DNA-grade sterile water (Fisher BioReagents, Pittsburgh, USA), 2.5 µL of 5X Green GoTaq® reaction buffer (Promega©, Madison, USA), 2.5 µmol dNTP mix, 0.002 mg/µL BSA, 2.5 µmol NL1 primer, 2.5 µmol NL4 primer, 12.5 µmol MgCl2, 0.65 units GoTaq® DNA Polymerase and 1.5 µL of genomic yeast DNA. Amplification was performed in an Applied Biosystems® Veriti® 96-Well Fast Thermal Cycler (Foster City, USA) using the following parameters: 94°C for 3 min (1 cycle), 94°C for 40 s, 54°C for 40 s, 72°C for 50 s (32 cycles), 72°C for 10 min (1 cycle). Amplification was verified by visualizing a 1% agarose gel using a Gel Logic 400 Imaging System. Samples were ‘cleaned’ prior to submission by adding 1 µL EXO-sapIT enzyme to 5 µL of post-PCR product and placed in the thermocycler under the following parameters: 37°C for 15 min and 80°C for 15 min. To this reaction, 5 µmol NL1 primer was added before submitting to FADSS for one-way DNA sequencing using an ABI 3130 XL Genetic Analyzer. Completed sequences were viewed using Sequencher® software and unmatched bases were manually input into the sequence. These corrected sequences were 	 19	compared to the BLAST database (NCBI, 2014); species that had >97% similarity were considered a match.  2.2.3  Culture-independent yeast identification 2.2.3.i   DNA extraction Yeast DNA extractions were performed using the E.Z.N.A ® Stool DNA Kit (Omega Bio-tek Inc., Norcross, GA, USA) as per the manufacturer’s protocol with the slight modification of vortexing cells with 0.1mm glass beads in the provided extraction buffer for 30 min.  Extracted DNA samples from the same tank and stage were pooled for Illumina analyses. 2.2.3.ii  Illumina MiSeq library preparation  Extracted DNA was subjected to Illumina MiSeq library preparation in two PCR-based steps (IBEST Genomics Resource Core, Idaho, USA). First, sample DNA was amplified using the forward primer CS1-BITS (5'-CTACCTGCGGARGGATCA-3’) and reverse primer CS2-B58S3 (5’-GAGATCCRTTGYTRAAAGTT-3’) as created by Bokulich and Mills (2013) in a PCR reaction mix containing 10 µL 10x PCR buffer (Promega), 5 µL 25 mM MgCl2, 0.6 µL 20 mg/mL BSA, 1 µL 10mM dNTPs, 1.5 µL each of forward and reverse primers, 0.52 µL GoTaq® DNA polymerase, 1 µL sample DNA and topped with DNA-grade sterile water for a total reaction volume of 50 µL. Reactions were amplified in an Applied Biosystems® Veriti® 96-Well Fast Thermal Cycler (Foster City, USA) under the following cycle parameters: 95°C for 2 min, 15 cycles of 95°C for 40 s, 55°C for 40 s, and 68°C for 1 min, and one cycle of 68°C for 5 min. PCR amplification was confirmed by running 5 µL of PCR product on a 1.5% agarose gel containing 1x SYBR® Safe DNA gel stain (Life Technologies, Carlsbad, USA) and viewed under UV transillumination via the Gel Logic 400 Imaging System (Mandel, Rochester, USA). 	 20	Samples were diluted 5 or 10-fold based on band strength on the resulting gel. Unique 8 nucleotide P5/P7 barcode sequences (Illumina, Inc., San Diego, USA) were attached to each amplified sample under the following reaction conditions: 2 µL of 10x PCR Buffer (Promega), 3.6 µL of 25 mM MgCl2, 0.6 µL of 20 mg/mL BSA, 0.4 µL of 10 mM dNTPs, 0.75 µL of both P5 and P7 barcode sequences, 0.2 µL GoTaq® DNA Polymerase and 1 µL PCR product topped with DNA-grade sterile water to a final volume of 20 µL. Reactions were amplified under the following cycle parameters: 95°C for 1 min, 10 cycles of 95°C for 30 s, 60°C for 30 s and 68°C for 1 min, then one cycle of 68°C for 5 min. Amplicons were again visualized on a 1.5% agarose gel under UV transillumination to confirm successful barcode attachment.  A mock community consisting of known extracted DNA ratios of pure Saccharomyces cerevisiae, Saccharomyces uvarum, Hanseniaspora uvarum and Torulospora delbruekii cultures was created. Single isolates of each yeast species were placed in 2 mL of liquid YEPD media and grown at 28°C for 2 hours and DNA was extracted as per Section 2.2.3.i. DNA concentration of each yeast was determined by subjecting 1 µL of extracted DNA to a Nanodrop (Fisher, San Diego, USA). Extracted DNA was then pooled, and subjected to Illumina library preparation in the same manner as winery samples to determine detection accuracy within sequencing results.  Samples were submitted to the IBEST Genomics Resources Core facility at the University of Idaho, USA for quantification, normalization, pooling and 300 base pair (bp) paired-end sequencing on the Illumina MiSeq Desktop Sequencer (Illumina Inc., San Diego, USA).      	 21	2.2.4  Data analysis  2.2.4.i   Fermentation kinetics and diversity  Student’s t-tests were conducted to determine whether CFU/mL differed between the two treatments of each varietal at all four stages of fermentation, and deemed significantly different at p < 0.05 (JMP® Version 11. SAS Institute Inc., Cary, NC, 1989-2015). Simpson’s Index of Diversity was used to determine yeast species and strain diversity for each tank and barrel at each sampling stage or both culture dependent and independent analyses (Simpson, 1949). Culture-independent data were first transformed such that the relative abundance of organism sequences (expressed as a decimal) were multiplied by 100. Two-way ANOVA was conducted to determine if any differences in species diversity occurred between each treatment at all stages of fermentation (JMP®, Version 11. SAS Institute Inc., Cary, NC, 1989-2015). 2.2.4.ii  Illumina sequence analysis Raw fastq paired end reads were merged using Fast Length Adjustment of Short reads (FLASh) software as per Magoc and Salzberg (2011) and trimmed at both ends using the extended randomized numerical aligner (ERNE-filter) to remove low quality or contaminated reads. The 300 bp reads were quality filtered to remove sequences with a quality score of <Q20, any barcode/primer errors and any sequences containing <75% of total read length (Bokulich et al. 2013b). These filtered sequences were demultiplexed and assigned operational taxonomic units (OTUs) to fungal sequences of 99% similarity or higher using Quantitative Insights into Microbial Ecology (QIIME), the bioinformatics pipeline for performing microbiome analyses of DNA sequencing data. OTUs were classified taxonomically against the UNITE database which identified the fungal species represented by the DNA sequences. Any unassigned sequences were 	 22	removed prior to analyses and all identified species present above 0.01% relative abundance were included in subsequent Simpson’s index of diversity, NMS, and PERMANOVA analyses. 2.2.4.iii Community structure  Sequence read data and culture-dependent yeast isolate data were transformed to represent relative abundances of species detected throughout fermentation treatments in all stages and fermentation vessels for both microsatellite/sequencing and Illumina analyses. Yeast isolate data per sample were transformed to reflect the relative abundance in decimal form, and distance matrices between yeast communities were created using a Bray-Curtis dissimilarity index (Bray and Curtis, 1957). Communities were visualized by non-parametric multidimensional scaling (NMS) using PC-ORD™ (Version 6, MJM Software Design). Untransformed species abundance data were subjected to PERMANOVA analysis using the Bray-Curtis dissimilarity index of the species abundance modelling the treatment and stage as fixed factors (PRIMER 7+, PRIMER-E Ltd., Devon, UK).    	 23	2.3 Results 2.3.1 Fermentation kinetics of inoculated and spontaneous Pinot noir and Chardonnay Detailed accounts of the fermentation parameters (sample dates, vessel data, and vineyard information) for all study vessels are found in Tables A1-4 of Appendix A. Comparing fermentation kinetics in terms of CFU/mL and °Brix at each stage between spontaneous and inoculated fermentations of Pinot noir and Chardonnay revealed many differences (Figure 2.1A and B). Significantly higher CFU/mL were found in inoculated Pinot noir at cold soak (p < 0.001), and in spontaneous Pinot noir during the early stage (p = 0.007); however, no difference in CFU/mL was indicated between mid and late stages of Pinot noir treatments (p = 0.256, p = 0.342). During Chardonnay fermentations CFU/mL were only significantly higher in the mid stage of spontaneous fermentation (p < 0.001). No differences in °Brix were found between treatments for both varietals at cold soak, early, mid and late stage, respectively (Pinot noir: p = 0.781, p = 0.355, p = 0.380, p = 0.779; Chardonnay: p = 0.890, p = 0.724, p = 0.882, p = 0.768). 	 24	 A)     B)   Figure 2.1. Fermentation kinetics of inoculated (I) and spontaneous (S) fermentations of A) Pinot noir and B) Chardonnay (n=3). Colony forming units per milliliter (CFU/mL) and °Brix was measured and compared at each stage of fermentation.   2.3.2 Yeast succession throughout Pinot noir and Chardonnay fermentations Microsatellite and sequencing showed that many S. cerevisiae and non-Saccharomyces yeasts were present during fermentation (Figure 2.2). For all fermentations, non-Saccharomyces yeasts were the only isolates identified during cold soak. As fermentation progressed to early stage of inoculated Pinot noir (Figure 2.2A), non-Saccharomyces yeasts decreased sharply while S. cerevisiae isolates were most prominent. Subsequently, non-Saccharomyces isolates increased 	 25	in abundance during both mid and late stages. Similar to the inoculated Pinot noir, S. cerevisiae isolates became dominant over non-Saccharomyces at early stage of spontaneous Pinot noir (Figure 2.2B); however, S. cerevisiae became the only yeast isolates present during mid and late stages.  Inoculated and spontaneous Chardonnay fermentations showed nearly identical trends in S. cerevisiae and non-Saccharomyces isolates at each stage (Figure 2.2 C and D). Early stage was nearly devoid of non-Saccharomyces isolates, before increasing to their highest number at mid stage either equaling or dominating over the S. cerevisiae isolates. Finally, non-Saccharomyces isolates showed a slight decrease in prevalence during late stage while S. cerevisiae isolates increased in abundance. 	 26	 A)                                                        B) Cold SoakEarly MidLate020406080100Relative abundance (%)Cold SoakEarly MidLate020406080100Relative abundance (%)Non-Saccharomyces speciesSaccharomyces cerevisiae      C)                                                        D)              Cold SoakEarly MidLate020406080100Relative abundance (%)Cold SoakEarly MidLate020406080100Relative abundance (%)	 Figure 2.2. Relative abundances of Saccharomyces cerevisiae and non-Saccharomyces isolates identified using microsatellite and sequencing analyses within A) inoculated and B) spontaneous Pinot noir fermentations, and C) inoculated and D) spontaneous Chardonnay fermentations (n=3).   	 27	2.3.3 Yeast strain and species identities found by culture dependent analyses Many yeast species and strains were isolated from all fermentations (Figure 2.3). Specifically, commercial strains, (shown in bold) used concurrently or previously by the winery as inocula were detected in all Pinot noir and Chardonnay vessels during early, mid, and late fermentation stages. Hanseniaspora uvarum dominated cold soak of all fermentations, before being succeeded by a variety of other non-Saccharomyces species, and commercial and unknown S. cerevisiae strains. The Pinot noir inoculant strain RC212 implantation was 34.7% in early, persisting to 52.8 % in mid, and 44.4% in late stages of the inoculated fermentations, respectively (Figure 2.3A). The low percentage of the inoculum strain allowed for a relatively high richness of other commercial strains, which resembled more of a spontaneous fermentation. A possible reason for the low implantation, is the rather atypical increase in relative abundance of Pichia occidentalis during early (13.9%), mid (15.3%), and late (36.1%) stages. The spontaneous fermentations of Pinot noir were also dominated by commercial strains at all stages, the most prominent being Lalvin D254 present at 47.2% at early, 45.8% at mid, and 56.9% at late stage, respectively. This strain was previously used as inoculum in the winery; however, it was removed from use following the 2011 vintage. (Figure 2.3B). Furthermore, many unknown S. cerevisiae strains were identified composing up to 18.1% of the isolate abundance. The inoculant Zymaflore VL2 implanted at 9.8% during early, and persisted to 52.8% at mid, and 54.2% at late stages of the inoculated Chardonnay fermentations (Figure 2.3C). Similar to the inoculated Pinot noir, the relatively low persistence of the inoculum resulted in a high diversity of commercial strains during early stage, more resembling a spontaneous fermentation. Furthermore, Lalvin D254 was found to be present at 47.2% relative abundance at early stage, indicating the persistence of this strain; however, the inoculum increased to 54.2% by late stage 	 28	more reflecting an inoculated fermentation except that Pichia occidentalis had increased to an average of 34.7% relative abundance. The early stage of the spontaneous Chardonnay fermentation being rich in commercial yeast strains was typical of this fermentation type and in fact was nearly identical to the yeast communities found during the inoculated fermentations (Figure 2.3 D). However, the mid and late stages were more typical of an inoculated fermentation with the inoculum strain VL2 present at 8.3% at early, 23.6% at mid and 62.6% at late stage, only allowing Pichia occidentalis to increase to 37.4% relative abundance by late stage.  	 29	     A)         B)       Cold soakEarly Mid LateCold soakEarly Mid Late020406080100StageRelative abundance (%)H. uvarumH. meyeriH. opuntiaeC. albicansC. diversaC. californicaLalvin D254Lalvin RC212Lalvin CY3079Vitilevure 3001Vitilevure 58W3Fermol Arome PlusFermol BlancZymaflore VL2Zymaflore VL3Rhone 4600Redstart MontrachetLallemand BM45Unknown 1Unknown 2Unknown 3Unknown 4Unknown 5Unknown 6Unknown 7Unknown 8Unknown 9Unknown 10Unknown 11P. kluyveriP. occidentalis                C)                            D) Cold soakEarlyMid LateCold soakEarlyMid Late020406080100StageRelative abundance (%)H. uvarumP. occidentalisP. sporocuriosaC. zemplininaS. uvarumLalvin D254Lalvin RC212Lalvin CY3079Fermol Arome PlusZymaflore VL2Rhone 4600Unknown 1Unknown 2Unknown 3Unknown 4Unknown 5Unknown 7Unknown 9Unknown 12  Figure 2.3. Average strain and species abundance as determined by microsatellite and sequencing analyses during all fermentation stages of A) Pinot noir inoculated with Lalvin RC212 and B) spontaneous Pinot noir, and C) Chardonnay inoculated with Zymaflore VL2 and D) spontaneous Chardonnay (n=3). All yeast strains used as commercial inocula concurrently or previously by the winery are shown in bold. Cold soakEarly Mid LateCold soakEarly Mid Late020406080100StageRelative abundance (%)H. uvarumH. meyeriH. opuntiaeC. albicansC. diversaC. californicaLalvin D254Lalvin RC212Lalvin CY3079Vitilevure 3001Vitilevure 58W3Fermol Arome PlusFermol BlancZymaflore VL2Zymaflore VL3Rhone 4600Redstart MontrachetLallemand BM45Unknown 1Unknown 2Unknown 3Unknown 4Unknown 5Unknown 6Unknown 7Unknown 8Unknown 9Unknown 10Unknown 11P. kluyveriP. occidentalisCold soakEarlyMid LateCold soakEarlyMid Late020406080100StageRelative abundance (%)H. uvarumP. occidentalisP. sporocuriosaC. zemplininaS. uvarumLalvin D254Lalvin RC212Lalvin CY3079Fermol Arome PlusZymaflore VL2Rhone 4600Unknown 1Unknown 2Unknown 3Unknown 4Unknown 5Unknown 7Unknown 9Unknown 12Cold soakEarlyMid LateCold soakEarlyMid Late020406080100StageRelative abundance (%)H. uvarumP. occidentalisP. sporocuriosaC. zemplininaS. uvarumLalvin D254Lalvin RC212Lalvin CY3079Fermol A ome PlusZymaflore VL2Rhone 4600Unknown 1Unknown 2Unknown 3Unknown 4Unknown 5Unknown 7Unknown 9Unknown 12Cold soakEarly Mid LateCold soakEarly Mid Late020406080100StageRelative abundance (%)H. uvarumH. eyeriH. opuntiaeC. albicansC. diversaC. californicaLalvin D254Lalvin RC212Lalvin CY3079Vitilevure 3001Vitilevure 58 3Fer ol Aro e PlusFer ol BlancZy aflore VL2Zy aflore VL3Rhone 4600Redstart MontrachetLallemand BM45Unknown 1Unknown 2Unknown 3Unknown 4Unknown 5Unknown 6Unknown 7Unknown 8Unknown 9Unknown 10Unknown 11P. kluyveriP. occidentalis . i t s . al i  P. kluyveri H. meyeri H. opi iae C. diversa C. calif i  Cold soakEarlyMid LateCold soakEarlyMid Late0204060801 0StageRelative abundance (%)H. uvarumP. o cidentalisP. sporocuriosaC. zemplininaS. uvarumLalvin D254Lalvin RC212Lalvin CY3079Fermol Arome PlusZymaflore VL2Rhone 46 0Unknown 1Unknown 2Unknown 3Unknown 4Unknown 5Unknown 7Unknown 9Unknown 12. uvarum . occi e tali  . s ocuri  . z li i  S. uvaru  i  C212 i  254 tilevure 3 01 tilevure 58W3 l rome Plus   achet ll a  M45 l e L2 l i  Y3079 Fer lanc l re L3   Cold soakEarly Mid LateCold soakEarly Mid LateRelative abundance (%)Unkno n 1 Unkno n 2 Unkno n 3 Unkno n 4 Unkno n 5 Unkno n 6 Unknown 7 Unknown 8 Unknown 9 Unknown 10 Unknown 1    alvin 254 alvi  212   r l  l s fl e  hone 4600   Cold soakEarlyMid LateCold soakEarlyMid LatetRelative abundance (%). r. i t lir ili i. rl i  l i  l i  r l r  lfl r                       	 30	2.3.4 Culture dependent and independent species richness and community structure  The expected mock community relative abundance based on known extracted DNA concentrations of S. cerevisiae, S. uvarum, T. delbruekii, and H. uvarum and the resulting community as per Illumina sequencing are shown in Figure 2.4. Relative abundances of H. uvarum and T. delbruekii were higher while S. cerevisiae and S. uvarum were lower than expected; however, community composition remained relatively similar. Illumina was able to detect other organisms (each present at <1% relative abundance), however 97.6% of the sequences returned belonged to the mock community yeast species. Expected Observed050100Mock communitiesRelative abundance (%)Saccharomyces cerevisiaeTorulaspora delbrueckiiHanseniaspora uvarumOtherSaccharomyces uvarum Figure 2.4. Expected and observed relative abundances of the yeast species H. uvarum, T. delbruekii, S. cerevisiae, and S. uvarum present within the mock community.  Culture dependent (microsatellite and sequencing) and independent (Illumina MiSeq) identification of the yeast communities (Appendix B, Tables B1-4) within the inoculated and spontaneous fermentations of Pinot noir and Chardonnay revealed many similarities when considering the dominant yeast species present above 1% relative abundance (Figure 2.5). Aside 	 31	from cold soak where only H. uvarum was detected by both methods, dominant species richness of early, mid, and late stage was quite similar regardless of the identification method used. In the inoculated fermentations of Pinot noir (Figure 2.5 A), the dominant species S. cerevisiae, H. uvarum and P. occidentalis were detected by both methods in early, mid, and late stages, despite having different relative abundances. Conversely, in the spontaneous fermentations of Pinot noir (Figure 2.5 B), only the dominant species S. cerevisiae was detected by both methods in early, mid, and late stages, again reporting different relative abundances. The inoculated and spontaneous fermentations of Chardonnay were composed of the same dominant species S. cerevisiae and P. occidentalis as determined by both culture dependent and independent methods (Figure 2.5 C and D); however, their relative abundances were vastly different. In the culture dependent community, P. occidentalis was thought to be absent in early, dominant in mid, and reached a considerable relative abundance of 34.7% and 37.4% in the late stage of inoculated and spontaneous fermentations, respectively. In the culture independent community, P. occidentalis increased in abundance as the stages of fermentation progressed for both inoculated and spontaneous treatments, but never reached higher than 15% relative abundance.   	 32	Cold soak Early Mid Late020406080100StageAverage relative abundance (%)Cold soak Early Mid Late020406080100StageAverage relative abundance (%)Cold soak Early Mid Late020406080100StageAverage relative abundance (%)Cold soak Early Mid Late020406080100StageAverage relative abundance (%)Cold soak Early Mid Late020406080100StageAverage relative abundance (%)Cold soak Early Mid Late020406080100StageAverage relative abundance (%)Cold soak Early Mid Late020406080100StageAverage relative abundance (%)Cold soak Early Mid Late020406080100StageAverage relative abundance (%)Ai) Aii)Bi) Bii)Cii)Ci)Di) Dii) Figure 2.5. Relative abundances of dominant species (greater than 1% relative abundance) present within all stages of A) inoculated Pinot noir, B) spontaneous Pinot noir, C) inoculated Chardonnay, and D) spontaneous Chardonnay as determined by culture dependent (i) and independent (ii) methods. H. uvarumS. cerevisiae P. occidentalisC. californicaC. albicansH. meyeriH. opiontiaeC. diversaC. zemplininaS. uvarumA. pullulansD. tassianaK. aerobiaB. carolinianaR. nothofagiSaccharomycetales spAscomycota spM. bainieri	 33	Yeast community structure was different between inoculated and spontaneous Pinot noir treatments regardless of the identification method employed. Both the culture dependent and independent methods showed that the inoculated and spontaneous fermentations contained significantly different communities (p = 0.015 and p = 0.041, respectively). Thus, the NMS ordinations showed distinct spatial separation between the two treatments (Figure 2.6 A). Conversely, NMS ordinations of Chardonnay showed no spatial separation (Figure 2.6 B) because yeast community structure was not found to be different between inoculated and spontaneous fermentations as determined by both the culture dependent or independent methods respectively (p = 0.517, p = 0.608, respectively). Ai)             ii)  Bi)              ii)   Figure 2.6. NMS ordinations of total yeast species composition based on Bray-Curtis dissimilarity matrices. Ordinations are reported for inoculated and spontaneous fermentations of A) Pinot noir and B) Chardonnay for both culture dependent (i) and independent (ii) methods with significant treatment differences occurring between Pinot noir treatments only.  	 34	2.3.5 Diversity of yeast species  Using both identification methods, similar trends were detected when comparing between inoculated and spontaneous fermentations of both Pinot noir and Chardonnay (Figure 2.7). During fermentations of Pinot noir (Figure 2.7A), species diversity of inoculated tanks was higher than spontaneous fermentations during cold soak only when the yeast community was identified by the culture-independent method (p < 0.001). Diversity of spontaneous fermentations was significantly higher than inoculated fermentations during early (p < 0.001), mid (p < 0.001). and late (p < 0.001) stages as determined by both identification methods. Yeast diversity was not significantly different between treatments at early (p = 0.199, p = 0.254), mid (p = 0.455, p = 0.123), or late (p = 0.786, p = 0.275) stages of Chardonnay fermentations as determined by both identification methods (Figure 2.7B). 	 35	 A i)                                                                          ii) Cold soak Early Mid Late0.00.20.40.60.8StageSimpsons Index of DiversityCold soak Early Mid Late0.00.20.40.60.8StageSimpsons Index of Diversity Inoculated Spontaneous B i)                                                                          ii) Early Mid Late0.00.20.40.6StageSimpsons Index of DiversityEarly Mid Late0.00.20.40.6StageSimpsons Index of Diversity Figure 2.7. Simpson’s Index of Diversity for all stages of fermentation between inoculated and spontaneous treatments of A) Pinot noir and B) Chardonnay as determined by culture-dependent (i) and culture-independent (ii) microbial identification methods. Significant differences in diversity were determined by a two-way ANOVA and indicated by *.       *	 *	 *	 *	 *	 *	*		 36	2.4  Discussion 2.4.1  Yeast dynamics during inoculated and spontaneous fermentation For both varietals, spontaneous fermentations had higher CFU/mL at early, mid, and late stage despite sugar content remaining the same between treatments. It has been shown that an increase in CFU/mL as inoculated and spontaneous fermentations (in tank or barrel) progress to mid stage is attributed to S. cerevisiae and any non-Saccharomyces species capable of persisting through low levels of alcohol production, metabolizing available sugars. This trend is thought to be amplified specifically in inoculated fermentations due to the addition of a commercial inoculum that grows aggressively, with non-Saccharomyces yeasts still represented in the total CFU/mL (Egli et al. 1998; Domizio et al. 2006; Mercado et al. 2007; Zott et al. 2008). Contrary to this; however, for both grape varietals, the CFU/mL of spontaneous fermentations were found to be higher than their inoculated counterparts at early, mid, and late stages (Figure 2.1), rejecting my hypothesis 1-1. Despite confirming inoculum rehydration to be successful, this result is likely due to the addition of sulphites to all inoculated grape musts by this winery. Sulfur addition has been shown to limit yeast survival and therefore may have prevented the inoculum from establishing a population larger than in spontaneous tanks (Henick-Kling et al. 1998; Egli et al. 1998). Therefore, spontaneous tanks free from sulphite addition may have been more favorable to higher yeast growth. Furthermore, spontaneous Pinot noir fermentation seemed to contradict trends seen by Torija et al. (2001), in which CFU/mL reached their highest point at mid stage before dropping at late. Instead, CFU/mL decreased steadily as fermentation progressed, suggesting that the community of yeasts quickly reached a maximum size of both S. cerevisiae and non-Saccharomyces yeasts during the early stage of fermentation where the environment was in a relatively sulphite-free state. It is possible that the yeasts could no longer 	 37	be sustained by the must conditions, resulting in a drop in the cell abundance from the mid to late stages. This result may largely be exclusive to the spontaneous fermentations of Pinot noir rather than Chardonnay due to the larger volume of grape must being acted on by the yeasts, resulting in an increase of available nutrients. 2.4.2 Non-Saccharomyces and S. cerevisiae succession during fermentation Non-Saccharomyces, specifically H. uvarum, dominated the cold soak; however, contrary to conventional fermentation kinetics, non-Saccharomyces species (eg., Pichia occidentalis) were found in all subsequent stages of fermentation for both varietals. This resulted in relatively high non-Saccharomyces abundances in the late stages of the inoculated fermentations of Pinot noir, and the mid and late stages of both the inoculated and spontaneous treatments of Chardonnay. The spontaneous fermentations of Pinot noir were the only ones that followed conventional yeast succession, where non-Saccharomyces were dominant yeasts in cold soak and S. cerevisiae took over the fermentation in the later stages. Thus, my second hypothesis is only partially accepted. because in most of my fermentations, non-Saccharomyces persisted to the later stages. Recently, there has been a surge of research suggesting that non-Saccharomyces yeasts grow well into the mid and late stages and are capable of influencing the chemical and sensory characteristics of wine. These findings challenge the conventional notion that S. cerevisiae are the dominant yeasts in the later stages (Hansen et al. 2001; Jolly et al. 2006; Diaz et al. 2013). In inoculated vessels, the introduction of non-Saccharomyces in later stages may be due to a relatively low S. cerevisiae abundance, as supported by the relatively low CFU/mL, which may not allow for competition against non-Saccharomyces, resulting in their establishment. Nevertheless, this does not explain why spontaneously fermented Pinot noir does not contain non-Saccharomyces in the late stages. The absence of non-Saccharomyces yeasts in 	 38	the spontaneously fermented Pinot noir may be due to the establishment of a large population of S. cerevisiae such as Lalvin D254, which has previously been shown to be highly competitive (Lange et al. 2014). It may also be that non-Saccharomyces yeasts influenced the other fermentations more strongly as they had lower CFU/mL and were more vulnerable to the establishment of other yeasts should they enter the must (Perez-Martin et al. 2014). Consequently, in agreement with several studies, the survival of these yeasts to the later stages of fermentation suggests they are becoming more alcohol tolerant, and makes them likely contributors to the chemical and sensory attributes imparted in the wines (Ciani and Macarelli 1998; Moreira et al. 2011; Medina et al. 2013; Beckner-Whitener et al. 2015).  2.4.3 Yeast species and commercial strain identification  The inoculated fermentations of both Pinot noir and Chardonnay were not fermented primarily by the inoculum in the manner indicative of this fermentation treatment. Rather, the inocula Lalvin RC212 and Zymaflore VL2 in inoculated Pinot noir and Chardonnay fermentations, respectively, did not dominate the grape must following inoculation and had low implantation in early stage. Instead, the early stages of both fermentations resembled more of a spontaneous fermentation, given that they were abundant in other commercial strains used by the winery. The inocula did persist into late stage of both Pinot noir and Chardonnay fermentations to become the most abundant S. cerevisiae strain present, however they were not present in the typical abundance that is usually expected with inoculated fermentations. This result has been found in similar yeast composition studies conducted by Barrajon et al. (2010) and Lange et al. (2014), suggesting that despite inoculation, if the yeast strain is not competitive enough to prevent low implantation (below 80%), it may not become the dominant yeast and may not 	 39	impact the grape must as per the winemaker’s design. Instead other yeast strains and species will take residence and produce a wine similar to that expected from a spontaneous fermentation.  The spontaneous fermentations of Pinot noir and Chardonnay contained commercial strains used previously (eg., Lalvin D254, Redstar Montrachet, Lalvin CY3079) and concurrently (Lalvin RC212, Vitilevure 3001, Fermol Arome Plus, Zymaflore VL2). These results partially support hypothesis 3, since the majority of S. cerevisiae strains identified were commercial inoculum used by the winery. These results are in agreement with those of many researchers who have found that spontaneous fermentations are likely performed by yeast strains acting as inoculum for other varietals simultaneously, as well as strains present on the grape and winery surfaces, or dispersed during various winery processes (Santamaria et al. 2005; Blanco et al. 2011; Barata et al. 2012; Lange et al. 2014). During early Chardonnay fermentations, and through the entirety of spontaneously fermented Pinot noir stages, the most abundant commercial strain was Lalvin D254 despite not being used by this winery in 2 years. This indicates that certain yeast strains may be highly robust and capable of surviving in the winery environment long after being discontinued, therefore winemakers must be exceptionally mindful when choosing inoculum, because they may not be able to get rid of it. Nevertheless, inoculated fermentations of Pinot noir and inoculated and spontaneous fermentations of Chardonnay contained a large abundance of non-Saccharomyces species and both varietals contained an abundance of unknown S. cerevisiae strains. As mentioned previously, the prevalence of non-Saccharomyces in later stages may be due to a relatively low implantation of the inoculum during inoculated fermentation or low initial S. cerevisiae infiltration of spontaneously fermenting must (Barrajon et al. 2010), and the ability of non-Saccharomyces to survive in a higher alcohol content than once thought (Diaz et al. 2013). This may be due to the high 	 40	diversity of yeast in the early stage preventing any one S. cerevisiae yeast strain (commercial or unknown) from becoming dominant and being highly competitive against the establishment of the non-Saccharomyces population. Nevertheless, the spontaneously fermented Chardonnay resembled more an inoculated fermentation by late stage. There was a low yeast diversity and the single strain Zymaflore VL2 became dominant over the non-Saccharomyces population. Furthermore, the unknown S. cerevisiae strain abundance in the inoculated fermentation of Chardonnay and the inoculated and spontaneous fermentations of Pinot noir may be due to many factors. The literature suggests that unknown strains may actually be hybrid strains, resulting from mutations, mating of commercial inocula strains, and evolution of the yeast genomes due to fermentation conditions (Mortimer and Polsinelli 1999; Puig et al. 2000). Also, as the microsatellite database used to identify yeast strains only consists of S. cerevisiae strains used by all wineries studied at the UBC Okanagan campus, it is likely an unknown may be a commercial yeast strain. Should the winery misreport their yeast history or a strain be introduced to the cellar, such as from insects and human traffic, other commercial strains may establish residence and influence fermentations.  2.4.4 Species richness, community structure, and diversity as per culture dependent and independent analyses The mock community indicated that detection of species known to be present was relatively accurate, and community results of all inoculated and spontaneous fermentations were reliable. The mock community was composed of four yeast species, all of which were identified by Illumina sequencing. Other species in a relative abundance of < 1% were also identified in this community, which may be due to contamination of the original yeast communities or were misrepresentation of DNA sequences during the QIIME analysis (Schirmer et al. 2015). Next 	 41	generation sequencing, specifically using the Illumina MiSeq platform, allows for all fungal DNA within a fermentation sample to be identified. Nevertheless, this method may identify yeast cells not living at the time of sampling, or misrepresent the yeasts conducting fermentation, making the creation of a mock community to verify detection accuracy crucial when performing this method (Figure 2.4). It has been seen that the relative abundances of species detected can vary widely when subjecting the community to Illumina sequencing (Bokulich et al 2013a; Schirmer et al. 2015). My results of the mock community support the idea that relative abundances of species can vary widely, but they also support the method’s ability to detect the organisms known to be present.  When considering the yeast communities identified by both culture dependent (microsatellite and sequencing analyses) and independent methods (Illumina sequencing), the similarities and differences between these methods were shown. Firstly, hypothesis 4 was supported in that the dominant species richness was very similar for early, mid, and late stages of both inoculated and spontaneous fermentations of Pinot noir and Chardonnay, though the relative abundances of these species were different between methods (Figure 2.5). Cold-soak samples did not follow this trend as many dominant species were detected from the culture independent method, while culture dependent identification resulted in a single species detected. While both methods identified H. uvarum in dominant relative abundance, it is most likely that the culture medium prevented the detection of the abundance of species identified by Illumina. The culture medium, which selects for yeast species may have excluded the other fungal species from growing successfully; however, as they are fungal species, they were capable of being identified by fungal primers of the Illumina analysis. Furthermore, it is not known whether the dominant species identified by Illumina were alive at the time of sampling, therefore the lack of growth 	 42	and identification on media plates may be attributed to dead specimens. Few studies to my knowledge look simultaneously into applying culture dependent and independent methods to the same yeast samples (Cocolin and Mills 2003), but those pertaining to the microbial ecology of other systems have found many different results in terms of species richness and relative abundance agreeing between these methods (Kisand and Wikner 2003; Rantsiou et al. 2005). Moreover, no comparisons of culture dependent and independent methods of studying yeast communities specifically using the Illumina platform have been conducted to my knowledge, suggesting that this method may be an accurate alternative for studying yeast species within the later fermentation stages. Conversely, hypothesis 4 was partially accepted. While community compositions of the culture dependent and independent methods differed, they produced similar NMS ordinations for Pinot noir and Chardonnay. This indicated that studying the yeast species composition with either identification method allowed for the yeast communities between treatments to be deemed different, suggesting either method would be trustworthy when comparing community composition. Furthermore, when comparing diversity of inoculated and spontaneous fermentations, the same result was determined by both culture dependent and independent analyses for Pinot noir and Chardonnay, supporting hypothesis 5. Significantly higher diversity was seen between the cold soak stage of inoculated and spontaneous fermentations when calculated based on culture independent yeast identification; however, spontaneous fermentations were higher in diversity during the fermentative early, mid, and late stages. Therefore, this result is in accordance with similar studies that found increased S. cerevisiae and non-Saccharomyces yeast diversity in spontaneous fermentation (Schutz and Gafner 1993; Sabate et al. 1998; Dimaro et al 2007). Given that either yeast species identification method 	 43	reported the same result, this suggests that while the species composition may differ, trends in diversity may be nearly identical, regardless of the method chosen.  	 44	Chapter 3 – Chemical and sensory profiles of spontaneous and inoculated Pinot noir and Chardonnay fermentations at a Canadian winery  3.1 Synopsis There are many chemicals in fermenting wines that are a direct result of yeast metabolism and many of them influence sensory profile, the aroma and flavour by mouth attributes. Thus, it is imperative to identify the yeast community present within inoculated and spontaneous fermentations such that any differences in yeast-derived compounds and sensory profiles can be attributed to the yeasts that are present during fermentation. Chemical analyses from small-scale fermentations that were co-inoculated with S. cerevisiae and non-Saccharomyces species have been conducted; however, little is known about how chemical profiles and the aroma and flavour by mouth attributes (sensory profile) are affected by yeast communities (both S. cerevisiae non-Saccharomyces) of spontaneous and inoculated fermentations conducted at a commercial scale (Plata et al. 2003; Comitini et al. 2011; Medina et al. 2013). Spontaneously produced wines are thought to add increased complexity due to an increase in yeast diversity; however, very little research confirms this notion (Varela et al. 2009; Saberi et al. 2014). Alternately, inoculated fermentations have been thought to produce predictable sensory attributes due to the presence of high inoculum abundance from the commercial yeast added (Fleet and Heard 1993). The objectives of this study were to determine if yeast-derived compounds and sensory profiles differed between inoculated and spontaneous fermentations of Pinot noir, which were fermented by significantly different yeast communities. The lack of yeast community differentiation between inoculated and spontaneous fermentations of Chardonnay precluded characterization of chemical and sensory differences. The compounds that were measured included acetaldehyde, 	 45	propanol, ethyl acetate, isobutanol, amyl alcohol, isoamyl alcohol, and isoamyl acetate (Swiegers et al. 2005; Ciani et al. 2010; Barrajon et al. 2011; Styger et al. 2011). These compounds were selected because they have all been reported as abundant metabolites produced by yeasts during fermentation and can actively influence sensory profiles (Bisson and Karpel 2010). Moreover, 23 and 20 attributes associated with the Pinot noir and Chardonnay wines respectively, were assessed between inoculated and spontaneous treatments.	 46	3.2 Materials and methods 3.2.1  Sampling protocol The same experimental design outlined in Section 2.2.1 was utilized for research reported in this chapter. A 50 mL sample was taken at late stage from all replicate Pinot noir tanks and Chardonnay barrels, and aliquoted into 10 2 mL glass vials stored at -80ºC for chemical analysis. At the same time, 6 L from each tank and barrel were sampled for sensory analysis. The samples were stored at 5°C in 2 L glass bottles. At 2 and 5 weeks after sampling, wines were racked to remove precipitated solids. Samples were bottled 5 weeks after being collected at late stage, and sensory analysis was conducted 2 months after bottling. 3.2.2  Chemical analysis of wines  3.2.2.i   Enological analysis of wines Assessment of titratable acidity for all wines was conducted in duplicate at QGEW as modified from the Compendium of International Methods of Wine and Must Analysis (International Organization of Vine and Wine, 2008). A 10 mL sample of each wine was added to 30 mL of distilled water and titrated with 0.1 M sodium hydroxide (NaOH) to a final pH of 8.2. As determined by QGEW, titratable acidity (g/L) of the sample was equal to 75% of the NaOH titer volume as per the following equation: Sample NaOH titre volume x 0.75 = TA in g/L    (1) Assessment of volatile acidity (VA) for all wines was conducted in duplicate at the QGEW laboratory modified from the Compendium of International Methods of Wine and Must Analysis (International Organization of Vine and Wine, 2008). In a cash still, 0.5 mL of hydrogen peroxide was added to 10 mL of each wine sample and boiled until approximately 100 mL of distillate was collected. Four drops of phenolphthalein were added to the distillate and 	 47	titrated with 0.01 M NaOH until a persistent faint pink colour remained in the flask. Distilled water was titrated in the same fashion to determine the volatile acidity present in the blank. In accordance with winery practices, the blank titer volume was subtracted from all sample titer volumes, with 6% of this value equaling the volatile acidity (g/L) of the sample as per the following equation: (Sample titre volume – blank titre volume) x 0.06 = VA in g/L  (2) Assessment of residual sugar (°Brix) for all wines was also conducted in duplicate at QGEW using winery protocols according to the Rebelein method as derived from the Compendium of International Methods of Wine and Must Analysis (International Organization of Vine and Wine, 2008). Chardonnay samples were diluted tenfold due to substantial sugar content and Pinot noir samples were decolorized before undergoing sugar determination. In a 125 mL Erlenmeyer flask, 2 mL of wine, 10 mL of CuSO4 and 5 mL of NaK tartrate were added along with a few boiling chips, and boiled for 90 s. Lastly, 10 mL of KI, 25% H2SO4 and starch indicator, respectively, were added before the reaction was titrated with Na2S2O3	until white in colour. Residual sugar content was determined as per the following equation:  (Blank titre volume – sample titre volume) x dilution factor = residual sugar in g/L    (3) The pH of each sample was conducted in the BRAES lab at UBC Okanagan using a calibrated Accumet XL150 pH meter (Fisher Scientific, Ottawa, Ontario), with values for each sample being the average of duplicate sub-samples.  Assessment of alcohol content for all wines was conducted at the Summerland Research and Development Center (SRDC) using a Dujardin-Salleron electronic ebulliometer (Laboratoires Dujardin-Salleron, Noizay, France). The ebulliometer was calibrated using 10% ethanol and was flushed with 100 mL of each wine and drained to eliminate carry over from the 	 48	previous sample. The alcohol content in % volume by volume was determined using the calibrated scale provided by Dujardin-Salleron and the resulting alcohol content was averaged for each sample. Analyses were performed in duplicate.  3.2.2.ii  GC-FID analysis and standard curve preparation Samples collected at late stage centrifuged at maximum speed for 30 s to remove solids suspended within the wine. These samples were subjected to 0.2 µm filtration and placed in duplicate airtight 2 mL glass vials for gas chromatography flame ionization detection (GC-FID) analysis. Gas chromatography analyses were performed on a Shimadzu GC-2010 gas chromatograph (Shimadzu Corporation, Japan) equipped with a split/splitless injector and flame ionization detector (FID). Compounds within the samples were separated by direct injection via the injection port of 2 µL of each sample into a Rtx-1301® capillary column of dimensions 60 m x 0.25 mm x 1.4 µm (Restek, Bellefonte, PA, USA). The GC operating parameters were as follows: initial oven temperature of 36°C held for 5 min, then raised 1°C/min to 45°C and then 40°C/min to a final temperature of 190°C. The injector and detector temperatures were 200°C and 250°C, respectively, using H2 carrier gas at a flow rate of 1.2 mL per min and a split ratio of 1:20. Standards of 50, 100, and 150 mg/L of the fermentation-derived compounds acetaldehyde, propanol, ethyl acetate, isobutanol, amyl alcohol, isoamyl alcohol, and isoamyl acetate were also created in duplicate 2 mL vials to produce standard curves (Figure 3.1). Peak areas of corresponding compounds in wine samples allowed for concentration determination through comparison to standard curves using the linear regression equations derived from these curves.   	 49	             Figure 3.1. Average peak areas of all standard compounds at 50, 100, and 150 mg/L and the resulting standard curve equations (n=6; 3 concentrations x 2 replicates).   3.2.3  Sensory analysis of wines 3.2.3.i   Racking and bottling of wines Prior to any malo-lactic fermentation, the 6 L samples of each tank and barrel containing 40 ppm SO2 were kept at 5ºC at SRDC in Summerland, British Columbia. Samples were racked and returned to clean 2 L Kimax bottles after two and five weeks of cold storage to remove 	 50	precipitated solids from the wine. The headspace of each bottle was sparged with gaseous N2 to remove air from the neck of the bottle before returning to cold storage. Following the completion of racking, samples were bottled in 750 mL green glass wine bottles, corked with true corks, and kept at 5ºC until sensory analyses.  3.2.3.ii  Sensory standard preparation Flavour, aroma and mouthfeel standards for Chardonnay and Pinot noir wines were created at the Sensory lab at SRDC in accordance with standards created previously by Guinard and Cliff (1987) as well as Schlosser et al. (2005). Appropriate aroma and flavour by mouth attributes for both varietals were first determined by a preliminary evaluation of all wines by members of the Sensory lab at SRDC. A complete list of appropriate aroma and flavour by mouth standards for Pinot noir and Chardonnay wines was presented to the panel of experts during a roundtable training session. Following training, a shorter list of 23 and 20 appropriate aroma and flavour by mouth attributes for the Pinot noir and Chardonnay samples respectively, was agreed upon by all roundtable panelists. Standards were presented 11 Pinot noir and 10 aroma, flavour, and mouthfeel standards with each evaluation as outlined in Table 3.1. 	 51	 Table 3.1. Sensory standards created for Pinot noir and Chardonnay aroma and flavour by mouth assessments. Aroma/flavour standard Volume/amount Brand/source Pinot noir- each in 100mL neutral SOLA brand red wine Cherry 50 mL juice + 20 µL essence Safeway® canned sweet cherry juice and IFF Naturals® Cherry essence Red berry fruit 15 g each  Western Family® frozen strawberries and raspberries Black berry fruit 30 mL each Western Family® frozen blueberries and blackberries, and Ribena® Current concentrate Jammy 10 g each Smuckers® Pure raspberry and Pure strawberry jam Prune 100 mL  Welch's® Prune nectar Vegetative ~5 g steeped overnight Tomato vine Dill 4 g steeped overnight Safeway organics® fresh dill weed Black pepper 2 peppercorns steeped overnight Sensory lab Earthy 22 mL + 30 mL Safeway® canned potato and canned mushroom juice Visual assessment of Pinot noir  Colour High and low red colour Munsell Color ® 5R(high) 2/6 and 2.5R(low) 5/14 Chardonnay- each in 100mL neutral SOLA brand white wine Citrus 2x2 cm square steeped overnight Fresh lemon rind Pome 1 g ascorbic acid + 30 mL each Sigma Aldrich® ascorbic acid, Fresh juiced Granny Smith apple and Safeway® canned pear juice Tropical 50 mL SunRype® Mango Passion fruit juice Peach 66 mL Safeway® canned peach juice Spice 2 cloves, 1/8 tsp allspice, half cinnamon stick Sensory lab cloves, No Name® allspice and cinnamon sticks Oak/Vanilla 0.315 mL Presidents Choice® vanilla extract Honey 5 mL No Name® liquid honey Floral 40 uL Sweet clover natural essence (Les produits Naturome-Bell Inc. NV, USA) Tastes- in 250mL H2O   Sweetness 10g Sigma Aldrich® D- glucose Acidity 0.85g  EMD Millipore® Tartaric acid Sensations not requiring standards Body/Mouthfeel   Length of Aftertaste       	 52	3.2.3.iii Aroma, flavour by mouth, and colour analysis of wines Evaluations occurred in individual evaluation booths in the Sensory lab of SRDC under red light using CompuSense® sensory software. Eight panelists who had participated in the roundtable training conducted a duplicate assessment of spontaneous and inoculated Chardonnay and Pinot noir. Panelists were presented water, a spittoon for expectoration, and all sensory standards previously agreed upon. All panelist responses were encrypted such that an ID number was assigned to each panelist to ensure anonymity. Spontaneous and inoculated samples of a single varietal were coded with three-digit random numbers. Each panelist was presented a random order of samples. Samples were individually presented to panelists and evaluated for aroma followed by flavour by mouth on a 100 unit intensity scale with comments added at the panelists’ discretion. Following the first assessment of each wine panelists took a break and were presented light refreshments in the Sensory lab to prevent fatigue of their senses prior to the second assessment.  Visual assessments were deemed inappropriate for Chardonnay samples, as perceptible colour differences could not be determined between fermentation types; however, Pinot noir samples had a variety of colour differences. Visual assessment of colour intensity was conducted in duplicate using Compusense® by all panelists in natural light conditions following the replicate Pinot noir evaluations. Each wine was presented in 50 mL samples in a clear 100 mm x 15 mm Petri dish against a white background and colour intensity was compared to two colour chips that represented a red colour lighter and darker than all samples (Table 3.1). All responses were collected and stored in the Sensory lab for future statistical analysis.  	 53	 3.2.4  Data analysis 3.2.4.i   Standard concentration curves Linear regression models to produce linear equations using known concentrations of each standard compound to represent each compound studied (JMP®, Version 11).  3.2.4.ii  GC-FID analysis Means and standard errors (SE) of chemical concentrations pertaining to inoculated and spontaneous fermentations of Pinot noir and Chardonnay were determined using GraphPad (GraphPad Software, San Diego, CA, USA). One-way ANOVA was conducted to verify replicate tanks, within each treatment, did not significantly differ in compound concentration (JMP®, Version 11). Significant differences between concentrations of fermentation-derived primary compounds between both fermentation treatments were determined using Student’s t-tests. Mean concentrations of fermentation-derived primary compounds were subject to principal component (PCA) and hierarchical cluster analyses using the Ward method via an R 3.1.3 platform with a JMP user interface (JMP®, Version 11). 3.2.4.iii Sensory analysis Three-way ANOVA with a mixed effect model (Appendix C, Tables C1 and C2) was conducted to evaluate sensory data by assessing the effect of judges, samples, and replications, and all two-way interactions (SAS© SAS Institute Inc., Cary, NC, 1989-2015). Judge performance was evaluated using judge x sample interaction plots. Despite such plots having been used in the literature for identifying inconsistent judges, in this research it was not possible to determine one (or a few) atypical judges during the assessments; therefore, all the data from the judges were used to calculate mean sensory scores and perform the subsequent statistical analyses.  	 54	Means and standard errors of sensory attribute mean scores pertaining to inoculated and spontaneous Pinot noir and Chardonnay samples were calculated using GraphPad Prism (GraphPad Software, San Diego, CA, USA). Significant differences in aroma and flavor by mouth attribute intensity between inoculated and spontaneous treatments of Pinot noir and Chardonnay were determined by Student’s t-tests using GraphPad Prism 6.0c (GraphPad Software, San Diego, CA, USA). Principal component analysis (PCA) of the correlation matrix was produced by mean attribute scores using JMP®. Hierarchical cluster analysis using the Ward’s method was conducted using SAS® and JMP® software, respectively (JMP®, Version 11). To meet the statistical assumptions of PCA as per Dettmar et al. (2015) required the number of samples (12) be greater than or equal to the number of attributes (8 for Pinot noir and 6 for Chardonnay). The mean sensory scores of black pepper, earthy, and vegetative aroma and flavor were averaged to simplify the aroma and flavour component of the same attribute as performed by Cliff et al. (2002). A reduction in the number of attributes was determined by selecting the most significant p-values generated by the Student’s t-tests. The 8 Pinot noir attributes with the most significant p-values between treatment wines and the 6 Chardonnay attributes with the lowest p-values despite being insignificant were included in this analysis.   	 55	3.3 Results  3.3.1 Enological and GC-FID analyses of wines The enological characteristics of inoculated and spontaneous fermentations of Pinot noir and Chardonnay were compared at the late stage of fermentation (Figure 3.2). Significant differences were only found in volatile acidity (p = 0.003) and residual sugar (p = 0.028) between inoculated and spontaneous Pinot noir fermentations. No significant differences in final titratable acidity, pH, or alcohol content were present between treatments of Pinot noir, and no significant enological differences were present between Chardonnay treatments.  Pinot noir Chardonnay0246810Titratable Acidityg/LInoculatedSpontaneousPinot noir Chardonnay01234pHInoculatedSpontaneousPinot noir Chardonnay0.00.20.40.6Volatile acidityg/LInoculatedSpontaneousA BC DE*Pinot noir Chardonnay-1012345BrixInoculatedSpontaneousPinot noir Chardonnay051015Alcohol (%v/v)SpontaneousInoculated*  Figure 3.2. A) Titratable acidity, B) volatile acidity, C) residual sugar (°Brix), D) pH, and E) alcohol content of late-stage inoculated and spontaneous Pinot noir and Chardonnay fermentations. Significant differences (p ≤ 0.05) in chemical attributes are indicated by *. 	 56	The GC-FID analysis of yeast-derived compounds between inoculated and spontaneous fermentations of Pinot noir and Chardonnay resulted in only a single significant difference (Table 3.2). Only isoamyl acetate was found to differ between Pinot noir treatments (p = 0.005), as it was not detected (ND) within spontaneous samples. Furthermore, no differences in compound concentrations were determined between Chardonnay treatments.  Table 3.2. Mean concentrations and standard error (SE) of primary yeast-derived compounds within inoculated and spontaneous fermentations of Pinot noir and Chardonnay as detected by GC-FID analysis. P-values determined by Student’s t-tests (n=6).  Compound Inoculated Pinot noir ± SE (mg/L)a Spontaneous Pinot noir ± SE (mg/L)a p-value Compound Inoculated Chardonnay ± SE (mg/L)a Spontaneous Chardonnay ± SE (mg/L)a p-value Acetaldehyde 101.4 ± 12.4 107.1 ± 11.7 0.305 Acetaldehyde 100.1 ± 5.6 103.2 ± 13.2 0.851 Propanol 48.7 ± 7.1 31.9 ± 11.0 0.303 Propanol 39.2 ± 9.9 47.5 ± 17.2 0.703 Ethyl acetate 41.6 ± 22.4 89.7 ± 31.1 0.214 Ethyl acetate 23.6 ± 12.1 31.0 ± 5.7 0.638 Isobutanol 140.4 ± 13.9 165.2 ± 19.0 0.368 Isobutanol 151.0 ± 36.2 139.9 ± 41.7 0.706 Amyl alcohol 155.7 ± 35.0 196.8 ± 18.8 0.406 Amyl alcohol 89.4 ± 18.1 92.1 ± 21.5 0.817 Isoamyl alcohol 132.7 ± 21.5 126.0 ± 31.3 0.907 Isoamyl alcohol 16.8 ± 5.9 22.6 ± 1.3 0.334 Isoamyl acetate  28.8 ± 6.1 ND 0.005 Isoamyl acetate 12.6 ± 5.4 8.9 ± 7.0 0.876 aOne-way ANOVA of chemical concentrations produced between replicate tanks not shown    	 57	3.3.2 PCA and cluster analyses of yeast-derived compounds Principal component analysis (PCA) of the yeast-derived compound concentrations within duplicate samples of inoculated and spontaneous fermentations of Pinot noir showed certain compounds were more associated with each treatment (Figure 3.3Ai). Two factors contained 93.3% of the variation of the 12 samples to reveal this trend. Inoculated Pinot noir wines had similar patterns of propanol and isoamyl acetate concentrations, while spontaneous wines were more associated with ethyl acetate, isobutanol and amyl alcohol concentrations. Furthermore, hierarchical cluster analysis of the yeast-derived compound average concentrations per tank using the Ward’s method confirmed the chemical profiles clustered by Pinot noir fermentation treatment (Figure 3.3Aii). Contrary to this, PCA and hierarchical cluster analysis of the yeast-derived compound concentrations for inoculated and spontaneous Chardonnay did not reveal any compounds more associated with either fermentation treatment or treatment clustering despite two factors explaining 78.1% of the variation within the samples (Figure 3.3Bi and 3.3Bii). 	 58	 Ai)        ii)                      Bi)        ii)          Figure 3.3. Principal component and hierarchical cluster analyses using the Ward’s method of yeast-derived compound concentrations within inoculated (I) and spontaneous (S) fermentations of A) Pinot noir (n=12) and B) Chardonnay (n=12). Tanks (T1-3) and replicates (R1-2) are denoted for all samples in PCA analyses, with the average of the two replicate samples per tank being utilized for cluster analysis.  	 59	3.3.3 Aroma, flavour by mouth, and colour attribute intensities The mean attribute intensity scores of all aroma, flavour by mouth, and colour attributes assessed within inoculated and spontaneous samples of Pinot noir and Chardonnay are shown in Table 3.3. The attributes with the lowest p-values as determined by Student’s t-tests were selected for subsequent analyses.  	 60	 Table 3.3. Mean attribute intensity scores and standard error (SE) for inoculated and spontaneous fermentations of Pinot noir and Chardonnay. Attributes were assessed in duplicate and rated on a 0-100 intensity scale by a panel of 8 experts, with significant attributes shown in bold (n=48; 8 experts x 3 wines x 2 replicates). Pinot noir Chardonnay  Inoculated Spontaneous   Inoculated Spontaneous  Attribute Mean ± SE Mean ± SE p-value Attribute Mean ± SE Mean ± SE p-value Cherry Aroma 33.6 ± 3 39.2 ± 1.8 0.34 Citrus Aroma 50.9 ± 0.7 55.9 ± 0.5 0.11  Red Berry Aroma 32.2 ± 3.1 40.1 ± 1.2 0.025 Pome Fruit Aroma 39.0 ± 1.2 38.4 ± 2.2 >.999 Black Berry Aroma 37.5 ± 1.5 37.6 ± 1.5 >0.99 Tropical Aroma 48.3 ± 1.3 51.2 ± 0.3 0.882 Jammy Aroma 43.7 ± 0.9 46.2 ± 1.9 0.99 Peach Aroma 53.9 ± 2.6 55.7 ± 0.9 0.999 Prune Aroma 37.7 ± 0.7 32.4 ± 1.2 0.46 Spice Aroma 22.7 ± 1.7 20.4 ± 0.9 0.989 Vegetative Aroma 47.6 ± 1.4 37.3 ± 0.9 <0.001 Oak/Vanilla Aroma 25.3 ± 2.0 21.3 ± 1.4 0.425 Dill Aroma 42.9 ± 2.9 34.9 ± 0.5 0.022 Honey Aroma 37.3 ± 0.7 36.9 ± 0.9 >.999 Black Pepper Aroma 42.0 ± 0.9 33.8 ± 0.6 0.016 Floral Aroma 31.9 ± 1.6 30.5 ± 1.8 >.999 Earthy Aroma 45.3 ± 2.2 34.6 ± 2.1 <0.001 Citrus Flavour 52.3 ± 1.4 52.2 ± 0.5 >.999 Cherry Flavour 37.8 ± 1.2 46.3 ± 1.2 0.009 Pome Fruit Flavour 38.2 ± 1.6 40.1 ± 0.6 0.998 Red Berry Flavour 37.9 ± 1.2 42.6 ± 0.9 0.678 Tropical Flavour 49.2 ±  1.3 49.6 ± 1.6 >.999 Black Berry Flavour 36.0 ± 0.7 39.5 ± 0.8 0.968 Peach Flavour 57.1 ± 1.0 57.9 ± 1.1 >.999 Jammy Flavour 36.9 ± 0.9 38.2 ± 0.8 >0.999 Spice Flavour 18.1 ± 0.6 17.2 ± 0.4 >.999 Prune Flavour 32.7 ± 0.1 28.8 ± 0.3 0.909 Oak/Vanilla Flavour 22.8 ± 1.3 20.6 ± 0.5 0.989 Vegetative Flavour 48.1 ± 1.7 32.9 ± 1.6 <0.001 Honey Flavour 41.6 ± 1.7 42.2 ± 0.8 >.999 Dill Flavour 38.4 ± 2.1 29.5 ± 0.4 0.0063 Floral Flavour 32.3 ± 1.6 29.4 ± 0.4 0.906 Black Pepper Flavour 47.4 ± 1.1 38.9 ± 0.8 0.0114 Sweetness 64.0 ± 1.0 65.1 ± 0.5 >.999 Earthy Flavour 40.5 ± 1.3 28.4 ± 1.8 <0.001 Acidity 41.4 ± 1.5 41.3 ± 0.9 >.999 Colour intensity 24.9 ± 2.5 65.5 ± 4.9 <0.001 Body/ Mouthfeel 58.9 ± 1.6 57.6 ± 0.6 >.999 Sweetness 20.6 ± 0.3 21.3 ± 1.3 >.999 Length of Aftertaste 62.1 ± 0.7 63.0 ± 0.6 >.999 Acidity 49.5 ± 1.9 46.9 ± 1.4 0.999     Body/ Mouthfeel 45.1 ± 1.7 51.7 ± 1.8 0.598     Length of Aftertaste 56.2 ± 0.8 58.1 ± 0.5 >0.999      	 61	3.3.4 Sensory attribute differentiation of inoculated and spontaneous fermentations Each sensory attribute was compared between treatments, illustrating the significantly higher intensities of vegetative aroma (p < 0.001) and flavour (p < 0.001), dill aroma (p = 0.022) and flavour (p = 0.006), black pepper aroma (p = 0.016) and flavour (p = 0.011), and earthy aroma (p < 0.001) and flavour (p < 0.001) in inoculated Pinot noir fermentations. Spontaneous fermentations of Pinot noir were significantly deeper in colour (p < 0.001) and had profiles consisting of greater red berry aroma (p = 0.025) and cherry flavour (p = 0.009) intensity (Figure 3.4A). No differences in attribute intensity were detected between sensory profiles of Chardonnay treatments (Figure 3.4B).    	 62	 A)                               *                                                               *                                                                                                                                                                                                                                                                                       *                      *                                                                                                         *        *                                                                                                                        *                       *                                                                                            *                     *                                                                                        *      B)                 Figure 3.4. Radar plots illustrating differences between mean intensity scores for all attributes associated with inoculated (black) and spontaneous (blue) fermentations of A) Pinot noir and B) Chardonnay. Significant differences (p ≤ 0.05) are illustrated by *. 	 63	Principal component and hierarchical cluster analysis using the Ward’s method were preformed and illustrated in Figure 3.5. Two principal components explained 93.7% of the total variation in this data, with 79.4% being described by a single principal component. This component revealed the attributes appear to show patterns of association such that sensory attributes clustered based on treatment (Figure 3.5A). Colour, cherry aroma, and red berry aroma were associated with spontaneous Pinot noir samples, whereas black pepper, earthy, and vegetative attributes were associated with inoculated Pinot noir samples. No significantly different attribute intensities were present between the Chardonnay treatment sensory profiles (Table 3.3) and no attribute associations or clustering effect based on treatment was indicated despite 67.1% of the variation in the data being explained by two factors (Figure 3.5B).  	 64	 Ai)            ii)                Bi)          ii)          Figure 3.5. Principal component and hierarchical cluster analysis using the Ward’s method of most significantly different sensory attribute intensities associated with sensory profiles of inoculated (I) and spontaneous (S) fermentations of A) Pinot noir (n=12) and B) Chardonnay (n=12). Tanks (T1-3) and replicates (R1-2) are denoted for all samples in PCA analyses, with the average of the two replicate samples per tank being utilized for cluster analysis.   	 65	3.4 Discussion   3.4.1 Chemical differentiation by yeast communities of inoculated and spontaneous fermentations  Following alcoholic fermentation, there were very few differences in the characteristics of inoculated and spontaneously fermented Pinot noir and Chardonnay wines (Figure 3.2). Only the volatile acidity (VA) and residual sugar of the Pinot noir wines differed, supporting hypothesis 1. These characteristics are largely influenced by the yeast community within the grape must, as yeasts can dramatically decrease VA production while simultaneous utilizing sugars while fermentation occurs (Henschke 1997; Swiegers et al. 2005; Hernandez-Orte et al. 2006). The lower volatile acidity of the inoculated must may be directly related to production associated with the inoculum strain Lalvin RC212. Specifically, this yeast has been noted by the manufacturer as a low producer of volatile acidity. Furthermore, RC212 is known to be a fast fermenter, therefore it may also be largely responsible for the difference in residual sugar present at late stage of Pinot noir fermentation. Late stage inoculated samples were taken 5 days post-inoculation, and showed no residual sugar left for the yeasts to metabolize. Meanwhile, late stage spontaneous samples, taken 10 days after crush, still showed an average of 1.3 °Brix, indicating all fermentable sugars were not yet consumed. As RC212 was only a minor contributor to the spontaneous yeast community, it may not have strongly influenced the rate of fermentation.   As no significant enological differences appear to exist between Chardonnay treatments, this is likely a reflection of the statistically similar yeast communities present between inoculated and spontaneously fermenting barrels. Section 2.4 suggested statistically similar communities were present in both treatments which likely accounts for the lack of any yeast-derived must variation.  	 66	 Both S. cerevisiae and non-Saccharomyces yeasts have also been shown to modulate the volatile chemical constituents of wine through a variety of metabolic pathways (Fleet 2003; Swiegers et al. 2005; Beckner Whitener et al. 2015); therefore, as the compounds studied are all known to be yeast-derived, differences in the yeast community would need to exist for chemical differences to occur. The yeast community was found to be different between inoculated and spontaneous Pinot noir treatments, (see section 2.4.3 and 2.4.4) and composed of different abundances of S. cerevisiae strains and non-Saccharomyces species. This should allow for differences in commonly produced chemicals, while the lack of yeast community differences between inoculated and spontaneous Chardonnay should prevent chemical differentiation. Many studies have been performed, which reveal the chemical contribution of the genera Hanseniaspora, Pichia and Saccharomyces to the volatile chemical profile of wines when fermenting individually, or by co-inoculation (King et al. 2013; Comitini et al. 2011; Medina et al. 2013). However, few studies have compared the production of these primary yeast-derived compounds between an inoculated and spontaneously derived yeast community (Domizio et al. 2006). Table 3.2 revealed that of the yeast-derived compounds identified, only isoamyl acetate was found to differ between inoculated and spontaneous Pinot noir, partially supporting hypothesis 2. This compound has been associated with yeast genera such as Pichia, Hanseniaspora, and Saccharomyces. When considering culture dependent identification, the yeast compositions, specifically the abundances of Pichia and Saccharomyces, were found to be quite different, suggesting that the difference in the production of this compound may be more reflective of the differences in community structure found using culture dependent methods; however, culture independent identification suggested these species were all present in both inoculated and spontaneous fermentations of Pinot noir in rather similar abundances. Also, there 	 67	are many other factors that may have influenced these results for both Pinot noir and Chardonnay. As differences in these compounds were determined in samples that had been previously frozen, it is likely that some of the volatile constituents within the must had volatilized out of solution and thus were not detected. Furthermore, there was a large amount of variation in the compound concentrations determined between replicate tanks and barrels. Had more replicate samples been included, it may have allowed for significant differentiation between other compounds. Lastly, there are many other yeast-derived compounds known to be in grape must (Swiegers et al. 2005; Styger et al. 2011; Beckner Whitener et al. 2015); therefore, to more accurately detect chemical differences it is probable that a larger number of the less prominent chemicals need to be identified between treatments.  Despite few significant differences, the yeast-derived chemical concentrations did show clustering and association with both treatments of Pinot noir (Figure 3.3). The production of propanol and isoamyl acetate were shown to associate with the inoculated samples indicating the yeast community of this treatment produced this pattern of concentrations of these compounds. Conversely, acetaldehyde, ethyl acetate, isobutanol and isoamyl alcohol production were shown to associate with spontaneous samples again indicating that the yeast community of the spontaneous fermentations was responsible for producing the concentrations detected of these compounds. As the yeast communities between inoculated and spontaneous Chardonnay were not found to be significantly different, the lack of chemical associations and clustering based on treatment further emphasized that community differences must be present for chemical differences to occur. Therefore, fermentation treatment and thus the significantly different yeast communities between them did influence the yeast-derived volatile chemical production, much 	 68	like similar studies conducted concerning yeast species and strain communities at the laboratory level (Mateos et al. 2006; Hyma et al. 2011; Saberi et al. 2012).  3.4.2 Sensory differentiation by yeast communities of inoculated and spontaneous fermentations Preliminary assessment was evaluated by a 3-way ANOVA and reported in Appendix C for Pinot noir and Chardonnay sensory analyses. The intensity of 23 aroma, flavour by mouth, and colour attributes was compared for all inoculated and spontaneous samples of Pinot noir and significant sensory differences were determined between the Pinot noir treatments (Table 3.3). As inoculated and spontaneously fermented Pinot noir were fermented by significantly different yeast communities, this suggests the difference in attribute intensities can be attributed to the different yeasts producing them. Conversely, no significant differences between attribute intensities of inoculated or spontaneously produced wines were determined between treatments. This further stresses the direct relationship between yeast community and the sensory profiles produced, as the statistically similar Chardonnay treatments were indistinguishable from a sensory attribute perspective.   Direct comparison of the attribute intensity scores determined between inoculated and spontaneous Pinot noir demonstrated many differences in attribute intensities of inoculated and spontaneous fermentations of Pinot noir (Figure 3.4). Inoculated samples had higher intensities of 8 of the 11 significantly different attributes, supporting hypothesis 3. These results oppose those found by Egli et al. (1998) that suggest inoculated wines had less dominantly detected attributes than spontaneous, which they attributed to the low abundance of non-Saccharomyces species. Instead, my results of increased relative abundance and lower diversity of S. cerevisiae strains within inoculated samples may have been more responsible for the production of 	 69	dominant attributes. Furthermore, the increased intensity of colour in spontaneous fermentations may be due to the longer total fermentation time (Appendix A), allowing for an increase in colour extraction due to prolonged contact with the skins.   Principal component (PCA) and hierarchical analysis showed an association with cherry aroma, red berry aroma, and colour intensity with spontaneously fermented Pinot noir samples (Figure 3.5). Conversely, black pepper, earthy, and vegetative attributes were associated with inoculated samples. This result supports my hypothesis 4, in that there were attributes present in intensities that clearly represented the specific treatment in which they were found. Furthermore, PCA of the Chardonnay attributes revealed no relationship between attributes and the fermentation treatment, accentuating the idea that without significantly different yeast communities fermenting the must, sensory differences would not occur. Studies have looked into the sensory profiles of either inoculated or spontaneous fermentations and the responsible yeast communities (Blanco et al. 2011; Medina et al. 2013); however, most are laboratory scale and few have directly compared spontaneous and inoculated treatments concurrently (Domizio et al. 2006). Despite the inoculated and spontaneously fermented Pinot noir yeast communities not reflecting the typical community structure indicative of these treatments, sensory differentiation will occur when the same grape varietal is fermented by different yeast communities.  	 70	Chapter 4 – Conclusion 4.1 Summary This study was conducted on commercial inoculated and spontaneous fermentations of Pinot noir and Chardonnay at an Okanagan winery. Specifically, this study had two objectives: (1) to understand the diversity and composition of the yeast communities within spontaneous and inoculated fermentation treatments when identified by culture-dependent and independent methods and (2) to determine if common yeast-derived chemicals and sensory attribute profiles differed between the fermentation treatments as a result of community differences. Previous research showed that non-Saccharomyces species rarely dominated in the late stages of fermentation, therefore they would not persist in large enough numbers to influence the chemical or sensory attributes of the wines in tandem with S. cerevisiae. However, my results showed an abundance of unexpected non-Saccharomyces yeast isolates in all stages of fermentation, prompting the need for further study.  The first hypothesis of objective one was refuted, finding instead that the CFU/mL of spontaneous fermentations for both Pinot noir and Chardonnay were higher than those of the inoculated treatments. Typically, inoculated fermentations were thought to have higher yeast cell concentrations given the addition of large amounts of the inoculum strain(s); however, possibly the addition of SO2 to the grape must prior to inoculation was detrimental to the S. cerevisiae strains once entering the must and prevented their establishment, as this was not present in spontaneous fermentations free from SO2 addition. Furthermore, the second hypothesis was also refuted, as the yeast isolates identified by microsatellite and sequencing analysis revealed non-Saccharomyces established persistent populations in all but spontaneous Pinot noir tanks. This challenges the long-standing concept 	 71	that non-Saccharomyces species are alcohol intolerant, being that they were isolated abundantly and cultured despite being isolated from fermentations as high as 13% alcohol content. This means that both inoculated and spontaneous fermentations are vulnerable to infiltration by non-Saccharomyces species that are capable of persisting to the end of fermentation, potentially metabolizing and significantly influencing the chemical and sensory characteristics. It was found that previously and concurrently used commercial ADY, and many unknown S. cerevisiae strains were also capable of infiltrating fermentations. Hypothesis 3 was partially supported in that many of the yeast strains isolated in spontaneous fermentations of Pinot noir and Chardonnay were commercial inocula; however, the inoculated fermentations behaved like spontaneous fermentations at the earlier stages, questioning the efficacy and implantation of the yeast inocula and suggesting the inoculated strain is quite vulnerable to competition by other yeasts. While some isolates identified were commercial strains not introduced purposely to the winery environment, the bulk of spontaneous fermentation was carried out by commercial strains rather than primarily indigenous ones, as is common in old world wine regions.  Recently, culture independent identification methods for determining microbial communities have been established, allowing yeast communities within fermentations to be identified directly from the sample. When directly comparing the yeast communities identified by microsatellite/sequencing analysis (culture dependent) and Illumina MiSeq sequencing (culture independent) many similarities between these methods were identified. Hypotheses 4 and 5 were supported following the comparison of these methods. Relative abundances of the species identified by both methods differed as expected, however the dominant species were identified by both methods. Furthermore, both methods confirmed the Pinot noir treatments consisted of significantly different yeast communities and that spontaneous fermentations were 	 72	higher in diversity. Therefore, both methods are strong when studying structure and diversity of the yeast community as a whole.  The simple chemical analysis of these spontaneous and inoculated wines revealed few differences present between inoculated and spontaneous fermentations of Pinot noir treatments and no differences whatsoever between inoculated and spontaneous fermentations of Chardonnay. This confirmed hypothesis 1 of objective two, as differences appeared most driven by the unique yeast communities, as yeasts are most capable of influencing the volatile acidity and residual sugars in the must. Only isoamyl acetate production was found to differ between chemical profiles of inoculated and spontaneous fermentations of Pinot noir where it was elevated in inoculated wines; however, PCA and cluster analysis revealed a pattern in the volatile chemical concentrations most typically occurring in the inoculated or spontaneous fermentations. This showed that despite not being significantly different, certain compound concentrations were representative of either treatment, supporting the hypothesis that different yeast communities would produce chemical profiles unique to the different treatments. This would likely be strengthened by increasing the number of compounds identified to further identify more nuanced chemical differences.  Sensory analysis of these wines revealed that very different aroma and flavour by mouth attributes were prevalent within the sensory profiles of inoculated and spontaneous Pinot noir only. Inoculated wines were significantly higher in 8 attributes, supporting hypothesis 3. This result is contrary to similar studies; therefore, the increase in dominant attributes may be more attributed to the lower diversity within inoculated tanks, rather than the specific yeast species themselves. PCA and cluster analysis also supported hypothesis 4, being that sensory profiles of inoculated and spontaneously fermented Pinot noir were more associated with specific attributes, 	 73	while no associations were seen within Chardonnay fermentations. This implies that the resulting sensory profiles were different only when significantly different yeast communities were found between the inoculated and spontaneous treatments conducted on the same varietal.  4.2 Novelty of the research  This research provides a novel contribution to the existing research in multiple ways. Firstly, it is the first study conducted in the Okanagan region specifically looking at the non-Saccharomyces species present between inoculated and spontaneous fermentations and its effect on chemistry and sensory attributes. Moreover, this study is the first, to my knowledge, to compare yeast populations and diversity throughout fermentation as identified by both culture-dependent and independent methods, identifying the shortcomings and benefits of both methods. Also, this study aimed to act as a pilot project in comparing yeast metabolomics between inoculated and spontaneous fermentation. Should inoculated and spontaneous fermentation treatments have vastly different yeast communities, yeast-derived primary compounds may be produced in different concentrations only if a thorough amount of metabolites is identified. Lastly, in the Okanagan region it is unknown how yeast communities within inoculated and spontaneous fermentations interact within grape must to influence the resulting chemical and sensory profile prior to malolactic fermentation. Thus, this study provides insight into how inoculation practices and the yeast communities inherent to the fermentation styles conducted in our region may cause entirely different wines to be produced by the same varietal.  4.3 Winery management implications   There are many practical applications for the research conducted within the industrial winery setting. Primarily, having understood the non-Saccharomyces and S. cerevisiae strains present within these fermentation treatments will allow winemakers to determine whether 	 74	inoculation with the strains used was a good investment. As inoculation did not implant fully for either Chardonnay or Pinot noir ferments, a large amount of fermentation was conducted by strains not intended to infiltrate that must. This suggests that the characteristics of that inoculum were not imparted into the must at the expected level and may cause winemakers to choose a more vigorous and competitive yeast strain. Also, yeast-derived compound production only slightly differed between inoculated and spontaneous wines, suggesting to winemakers that something resident to inoculated tanks was responsible for the increased production of 3-methylbutyl ethanoate. As this compound is known to associate with non-Saccharomyces species, winemakers may benefit from diligently monitoring its production as a potential indicator of non-Saccharomyces species abundance. Lastly, spontaneously fermented Pinot noir was found to consist of fewer dominant sensory attributes and lower attribute intensities than inoculated wines, preventing the most influential attributes from overpowering the detection of others; therefore, winemakers may benefit from producing a larger volume of spontaneously fermented wines in which consumers would experience a broader array of aroma and flavour-by-mouth attributes than their more dominant-attribute inoculated counterparts. 4.4 Suggestions for further research  The results of this study suggest many further experiments could be conducted to better optimize yeast identification methods and the specific contributions of yeasts within fermentations to the chemical and sensory attribute profiles of spontaneous and inoculated fermentations. For example, prevalent populations of unknown S. cerevisiae yeast strains and non-Saccharomyces yeasts were found within all fermentations of Chardonnay and Pinot noir, so in-lab studies of the isolated organisms could determine their chemical and sensory input within fermentation. It would be useful to directly compare any unknown strain microsatellite loci 	 75	within the commercial database created at UBCO to see if these strains may be hybrids of commercial inocula, mutated commercial inoculum, or true unique strains. These studies could also determine the metabolic capabilities of non-Saccharomyces species as alcohol content increases, and the specific fermentation-derived compounds they are producing within grape must can be identified on a yeast by yeast basis. Also, similar to S. cerevisiae strains, non-Saccharomyces species are thought to be strain specific, therefore, it would be beneficial to understanding whether unique strains of the non-Saccharomyces species are present within these fermentations.  Additionally, this study brought the usefulness and accuracy of culture-dependent and culture-independent yeast identification methods into question, as conducting both methods of yeast community identification provided quite different yeast relative abundances. Recently, it has been concluded that the compound propidium monoazide (PMA) can successfully prevent the detection of non-viable yeasts when using culture-independent Illumina MiSeq sequencing of the yeast DNA within an entire sample; therefore, treating all further samples with this compound would allow for culture-independent yeast community identification to reflect live organisms only and more accurately reflect yeast diversity.  Should a large population of non-Saccharomyces species coexist with many S. cerevisiae strains, it would be beneficial to increase the scope of the metabolomics study to include more fermentation-derived compounds so the profile of this type of yeast community is determined. This would more concretely direct hypotheses regarding the sensory attributes that should be detected within those wines.  Ultimately, despite these results suggesting that the more diverse yeast population within spontaneous Pinot noir wines resulted in a sensory attribute profile with fewer dominant 	 76	attributes, it would be beneficial to take this premise further to determine if there is a direct association between increases in diversity and sensory complexity. Creating an in-lab study involving many grape musts inoculated with different diversities of yeast strains and species and assessing the resulting wines for the number of attributes associated with each wine would allow for the ‘complexity’ argument to be approached. Lastly, this study only focuses on the flagship varietals of the winery, therefore it would be of great interest to see if non-Saccharomyces species persist in similar abundances across fermentations of different varietals. Similarly, it would be of interest to determine if non-Saccharomyces species show similar trends at multiple wineries should this trend not be unique to the single winery studied. As vintages can vary considerably in the yeast communities found in fermentations annually, it is important to deduce whether this non-Saccharomyces influence is persistent or whether the results are unique to this vintage.   	 77	References  Barata, A., M. Malfeito-Ferreira and V. Loureiro. 2012. The microbial ecology of wine grape berries. International Journal of Food Microbiology. 153: 243-259.  Barnett, A. J., M. A. Delaney, E. Jones, A. B. Magson and B. Winch. 1972. The numbers of yeasts associated with wine grapes of Bordeaux. Archives of Microbiology. 83: 52-55. Barrajon, N., M. Arevalo-Villena, J. Ubeda and A. Briones. 2010. Competition between spontaneous and commercial yeasts in winemaking: Study of possible factors involved. Applied Microbiology and Microbial Biotechnology, 1035–1041. Barrajon, N., A. Capece, M. Arevalo-Villena, A. Briones and P. Romano. 2011. Co-inoculation of different Saccharomyces cerevisiae strains and influence on volatile composition of wines. Food Microbiology 28: 1080-1086. Beckner-Whitener M. E., S. Carlin, D. Jacobson, D. Weighill, B. Divol, L. Conterno, M. Du Toit and W. Vrhovsek. 2015. Early fermentation volatile metabolite profile of non-Saccharomyces yeasts in red and white grape must: A targeted approach. LWT- Food Science and Technology 64:412-422. Bisson, L.F. and J.E. Karpel. 2010. Genetics of yeast impacting wine quality. Annual Review of Food Science and Technology 1: 139–162.  Blanco, P., I. Orriols and A. Losada. 2011. Survival of commercial yeasts in the winery environment and their prevalence during spontaneous fermentations. Journal of Industrial Microbiology & Biotechnology 38: 235–239.   Blanco, P., J. M. Miras-Avalos, E. Pereira and I. Orriols. 2013. Fermentative aroma compounds and sensory profiles of Godello an Albarino wines as influenced by Saccharomyces cerevisiae yeast strains. Journal of Science, Food and Agriculture 93: 2849-2857.  	 78	Bokulich, N. A, and D. A Mills. 2012. Next-generation approaches to the microbial ecology of food fermentations. BMB Reports 45: 377–389.  Bokulich, N. A, and D. A Mills. 2013. Improved selection of internal transcribed spacer-specific primers enables quantitative, ultra-high-throughput profiling of fungal communities. Applied and Environmental Microbiology 79: 2519–2526.  Bokulich, N.A, M. Ohta, P.M. Richardson, and D.A Mills. 2013a. Monitoring Seasonal Changes in Winery-Resident Microbiota. PloS One 8: 1–11.  Bokulich, N.A., S. Subramanian, J.J. Faith, D. Gevers, J.I. Gordon, R. Knight, D.A. Mills, and J.G. Caporaso. 2013b. Quality-filtering vastly improves diversity estimates from Illumina amplicon sequencing. Nature Methods 10: 57–60.  Camarasa, C., I. Sanchez, P. Brial, F. Bigey, and S. Dequin. 2011. Phenotypic landscape of Saccharomyces cerevisiae during wine fermentation: evidence for origin-dependent metabolic traits. PloS One 6: e25147.  Campo, E., V. Ferreira, A. Escudero, and J. Cacho. 2005. Prediction of the wine sensory properties related to grape variety from dynamic-headspace gas chromatography-olfactometry data. Journal of Agricultural and Food Chemistry 53: 5682-5690.  Capece, A., R. Romaniello, G. Siesto, R. Pietrafesa, C. Massari, C. Poeta, and P. Romano. 2010. Selection of indigenous Saccharomyces cerevisiae strains for Nero d’Avola wine and evaluation of selected starter implantation in pilot fermentation. International Journal of Food Microbiology 144: 187–192.  Ciani, M., and F. Maccarelli. 1998. Oenological properties of non-Saccharomyces yeasts associated with wine-making. World Journal of Microbiology and Biotechnology 14: 199–203.  	 79	Ciani, M., F. Comitini, I. Mannazzu, and P. Domizio. 2010. Controlled mixed culture fermentation: a new perspective on the use of non-Saccharomyces yeasts in winemaking. FEMS Yeast Research 10: 123–133.  Cocolin, L., and D. A. Mills. 2003. Wine yeast inhibition by sulfur dioxide: a comparison of culture-dependent and independent methods. American Journal of Enology and Viticulture 54: 125-130.  Comitini, F., M. Gobbi, P. Domizio, C. Romani, L. Lencioni, I. Mannazzu, and M. Ciani. 2011. Selected non-Saccharomyces wine yeasts in controlled multistarter fermentations with Saccharomyces cerevisiae. Food Microbiology 28: 873–882.  Clavijo, A., I.L. Calderon, and P. Paneque. 2010. Diversity of Saccharomyces and non-Saccharomyces yeasts in three red grape varieties cultured in the Serrania de Ronda (Spain) vine-growing region. International Journal of Food Microbiology 143: 241-245. Cliff, M., D. Yuksel, B. Girard and M. King. 2002. Characterization of Canadian ice wines by sensory and compositional analyses. American Journal of Enology and Viticulture 53: 46-53.  Cordero-Bueso, G., T. Arroyo, A. Serrano, J. Tello, I. Aporta, M.D. Vélez, and E. Valero. 2011. Influence of the farming system and vine variety on yeast communities associated with grape berries. International Journal of Food Microbiology 145: 132–139. Culleré, L., A. Escudero, J. Cacho, and V. Ferreira. 2004. Gas chromatography-olfactometry and chemical quantitative study of the aroma of six premium quality spanish aged red wines. Journal of Agricultural and Food Chemistry, 52: 1653-1660.  Dettmar, B., M. Plattner, and C. Vetter. 2015. Underdetermination in multivariate data analysis. 11th Pangborn Sensory Science Symposium 3-27 August 2015 Gothenburg, Sweden Poster Presentation P2.121.  	 80	Díaz, C., A.M. Molina, J. Nähring, and R. Fischer. 2013. Characterization and dynamic behavior of wild yeast during spontaneous wine fermentation in steel tanks and amphorae. BioMed Research International 2013: 1-13.  Dimaro, E., D. Ercolini, and S. Coppola. 2007. Yeast dynamics during spontaneous wine fermentation of the Catalanesca grape. International Journal of Food Microbiology 117: 201–210.  Domizio, P., L. Lencioni, M. Ciani, S. Di Blasi, C. Pontremolesi, and M. P. Sabatelli. 2006. Spontaneous and inoculated yeast population dynamics and their effect on the organoleptic characters of Vinsanto wine under different process conditions. International Journal of Food Microbiology 115: 281-289.  Egli, C.M., W.D. Edinger, C.M. Mitrakui, and T. Henick-Kling. 1998. Dynamics of indigenous yeast populations and their effect on the sensory character of Riesling and Chardonnay wines. Journal of Applied Microbiology 85: 779-789.  Esteve-Zarzoso, B., P. Manzanares, D. Ramón, and A. Querol. 1998. The role of non-Saccharomyces yeasts in industrial winemaking. International Microbiology 1: 143–148.  Field, D., and C. Wills. 1998. Abundant microsatellite polymorphism in Saccharomyces   cerevisiae, and the distributions of microsatellites in eight prokaryotes and S. cerevisiae result from strong mutation pressures and a variety of selective forces. Proceedings of the National Academy of Sciences 95: 1647-1652.   Fleet, G.H and G. M. Heard. 1993. Yeasts growth during fermentation. In: Fleet. G.H., ed. Wine Microbiology and Biotechnology. Philadelphia, PA: Harwood Academic, pp. 27-55.  Fleet, G.H. Yeast interactions and wine flavour. 2003. International Journal of Food Microbiology 86: 11-22.  	 81	Francis, I. L., and J. L. Newton. 2005. Determining wine aroma from compositional data. Australian Journal of Grape and Wine Research 11: 114-126.  Goldstein, D. and Schlötterer, C. 1999. Microsatellites: Evolution and applications. Oxford University Press, Oxford, UK.  Guinard, J. X., and M.A. Cliff. 1987. Descriptive analysis of Pinot noir wines from Carneros, Napa and Sonoma. American Journal of Enology and Viticulture 38: 211-215.  Guzzon, R., G. Widmann, L. Settanni, M. Malacarne, N. Francesca, and R. Larcher. 2011. Evolution of Yeast Populations during Different Biodynamic Winemaking Processes. South African Journal of Enology and Viticulture 32: 242–251.  Hall, B., D.M. Durall, and G. Stanley. 2011. Population dynamics of Saccharomyces cerevisiae during spontaneous fermentation at a British Columbia winery. American Journal of Enology and Viticulture 62: 66–72.  Hannequin, C., A Thierry, G.F Richard, G. Lecointre, H.V. Nguyen, C. Gaillardin, and B. Dujon. 2001. Microsatellite typing as a new tool for identification of Saccharomyces cerevisiae strains. American Society for Microbiology 39: 551-559.  Hansen, E. H., P. Nissen, P. Sommer, J. C. Nielsen, and N. Arneborg. 2001. The effect of oxygen on the survival of non-Saccharomyces yeasts during mixed culture fermentations of grape juice with Saccharomyces cerevisiae. Journal of Applied Microbiology 91: 541–547.  Henick-Kling, T., W. Edinger, P. Daniel, and P. Monk. 1998. Selective effects of sulfur dioxide and yeast starter culture addition on indigenous yeast populations and sensory characteristics of the wine. Journal of Applied Microbiology 84: 865–876.   	 82	Henschke, P., 1997. Wine yeast. In: Zimmerman, F.K., Entian, K.D. (Eds.), Yeast Sugar Metabolism, Biochemistry, Genetics, Bio- technology and Applications. Technomic Publishing, Lancaster, UK, pp. 527–560.   Hernandez-Orte, P., M. Bely, J. Cacho, and V. Ferreira. 2006. Impact of ammonium additions on volatile acidity, ethanol, and aromatic compound production by different Saccharomyces cerevisiae strains during fermentation in controlled synthetic media. Australian Journal of Grape and Wine Research 12: 150-160.  Hyma, K. E., S. M. Saerens, K. J. Verstrepen, and J. C. Fay. 2011. Divergence in wine characteristics produced by wild and domesticated strains of Saccharomyces cerevisiae. FEMS Yeast Research 11: 540-551.  International Organization of Vine and Wine (OIV). 2008. Compendium of international methods of wine and must analysis. Ed. 2008. Volume 1, pp. 1-504.   Jolly, N.P, O. P. H. Augustyn, and I. S. Pretorius. 2006. The role and use of non-saccharomyces yeasts in wine production. South African Journal of Enology and Viticulture 27: 15– 38.  King, E. S., R. L. Dunn, and H. Heymann. 2013. The influence of alcohol on the sensory perception of red wines. Food Quality and Preference 28: 235-243.  Kisand, V., and J. Wikner. 2003. Combining culture-dependent and independent methodologies for estimation of richness of estuarine bacterioplankton consuming riverine dissolved organic matter. Applied Environmental Microbiology 69: 3607-3616.  Lachance, M., H.M. Daniel, W. Meyer, G.S. Prasad, S. P. Gautam, and K. Boundy-Mills. 2003. The D1/D2 domain of the large-subunit rDNA of the yeast species Clavispora lusitaniae is unusually polymorphic. FEMS Yeast Research 4:253-258.   	 83	Lange, J.N., E. Faasse, M. Tantikachornkiat, F. S. Gustafsson, L.C. Halvorsen, A. Kluftinger, D. Ledderhof, and D. M. Durall. 2014. Implantation and persistence of yeast inoculum in Pinot noir fermentations at three Canadian wineries. International Journal of Food Microbiology 180: 56-61.   Legras, J., O. Ruh, D. Merdinoglu and F. Karst. 2005. Selection of hypervariable microsatellite loci for the characterization of Saccharomyces cerevisiae strains. International Journal of Food Microbiology 102: 73-83.  Legras, J., D. Merdinoglu, J.-M. Cornuet, and F. Karst. 2007. Bread, beer and wine: Saccharomyces cerevisiae diversity reflects human history. Molecular Ecology 16: 2091–2102. Le Jeune, C., C. Erny, C. Demuyter, and M. Lollier. 2006. Evolution of the population of Saccharomyces cerevisiae from grape to wine in a spontaneous fermentation. Food Microbiology 23: 709–16. Li, E., C. Liu, Y. Liu. 2012. Evaluation of yeast diversity during wine fermentations with direct inoculation and pied de cuve method at an industrial scale. Journal of Microbiological Biotechnology 22: 960-966.  Loureiro, V., and M. Malfeito-Ferreira. 2003. Spoilage yeasts in the wine industry. International Journal of Food Microbiology 86: 23–50.   Magoc, T., and S. Salzberg. 2011. FLASH: Fast length adjustment of short reads to improve genome assemblies. Bioinformatics 27: 21.  Mardis, E.R. 2008. Next-generation DNA sequencing methods. Annual review of genomics and human genetics 9: 387-402  	 84	Martini, A. 2003. Biotechnology of natural and winery-associated strains of Saccharomyces cerevisiae. International Microbiology: the official journal of the Spanish Society for Microbiology 6: 207–209.  Mateos, J. A. R., F. Perez-Navado, and M. R. Fernandez. 2006 Influence of Saccharomyces cerevisiae yeast strain on the major volatile compounds of wine. Enzyme and Microbial Technology 40: 151-157.  Medina, K., E. Boido, L. Fariña, O. Gioia, M.E. Gomez, M. Barquet, C. Gaggero, et al. 2013. Increased flavour diversity of Chardonnay wines by spontaneous fermentation and co-fermentation with Hanseniaspora vineae. Food Chemistry 141: 2513–2521.  Mercado, L., A. Dalcero, R. Masuelli, and M. Combina. 2007. Diversity of Saccharomyces strains on grapes and winery surfaces: analysis of their contribution to fermentative flora of Malbec wine from Mendoza (Argentina) during two consecutive years. Food Microbiology 24: 403–412.  Molina, A. M., J. H. Swiegers, C. Varela, I. S. Pretorius, and E. Agosin. 2007. Influence of wine fermentation temperature on the synthesis of yeast-derived volatile aroma compounds. Applied Microbiology and Biotechnology, 77: 675-687  Moreira, N., C. Pina, F. Mendes, J. A. Couto, T. Hogg, and I. Vasconcelos. 2011. Volatile compounds contribution of Hanseniaspora guilliermondii and Hanseniaspora uvarum during red wine vinifications. Food Control 22: 662–667.   Mortimer, R., and M. Polsinelli. 1999. On the origins of wine yeast. Research in Microbiology 150: 199–204.  Perez, M.A., F.J. Gallego, I. Martinez, and P. Hidalgo. 2001. Detection, distribution and selection of microsatellites (SSRs) in the genome of the yeast Saccharomyces   cerevisiae as molecular markers. Letters of Applied Microbiology 33: 461-66.  	 85	Perez-Martin, F., S. Sesena, M. Fernandez-Gonzalez, M. Arevalo, and M. Llanos Palop. 2014. Microbial communities in air and wine of a winery at two consecutive vintages. International Journal of Food Microbiology 190: 44-53.   Plata, C., C. Millan, J.C. Mauricio, and J.M. Ortega. 2003. Formation of ethyl acetate and isoamyl acetate by various species of wine yeasts. Food Microbiology 20: 217–224.  Puig, S, A. Querol, E. Barrio, J. Perez-Ortin. 2000. Mitotic recombination and genetic changes in Saccharomyces cerevisiae during wine fermentation. Applied and Environmental Microbiology 66: 2057-2061.  Querol, A., M.T. Fernandez-Espinar, M. del Olmo, and E. Barrio. 2003. Adaptive evolution of wine yeast. International Journal of Food Microbiology 86: 3–10.  Rankine, B. C., and B. Lloyd. 1963. Quantitative assessment of dominance of added yeast in wine fermentations. Journal of the Science of Food and Agriculture 14: 793-798.  Rantsiou K., R. Urso, L. Iacumin, C. Cantoni, P. Cattaneo, G. Comi, and L. Cocolin. 2005. Culture-dependent and independent methods to investigate the microbial ecology of Italian fermented sausages. Applied and Environmental Microbiology 71: 1977-1986.  Richards, K.D., M.R. Goddard, and R.C. Gardner. 2009. A database of microsatellite genotypes for Saccharomyces cerevisiae. Antonie van Leeuwenhoek 96: 355–359.  Romano, P., C. Fiore, M. Paraggio, M. Caruso, and A. Capece. 2003. Function of yeast species and strains in wine flavour. International Journal of Food Microbiology 86: 169–180.   Sabate, J., J. Cano, A. Querol, and J. M. Guillamon. 1998. Diversity of Saccharomyces strains in wine fermentations: analysis for two consecutive years. Letters in Applied Microbiology 26: 452-455.  	 86	Saberi, S., M.A. Cliff and H.J.J. van Vuuren. 2012. Impact of mixed S. cerevisiae strains on the production of volatiles and estimated sensory profiles of Chardonnay wines. Food Research International 48: 725-735  Saberi, S., M. A. Cliff, and H. J. J. Van Vuuren. 2014. Comparison of genetic and enological characteristics of new and existing S. cerevisiae strains for Chardonnay wine fermentations. Food Biotechnology 28: 195-215.  Saerens, S.M.G., F.R. Delvaux, K.J. Verstrepen and J.M. Thevelein. 2010. Production and biological function of volatile esters in Saccharomyces cerevisiae. J. Microbial Biotechnology 3: 165-177.   San Miguel, P. 2011. Next-generation sequencing and potential applications in fungal genomics. Methods of Molecular Biology 722: 52-60  Santamaría, P., P. Garijo, R. López, C. Tenorio, and A. R. Gutiérrez. 2005. Analysis of yeast population during spontaneous alcoholic fermentation: effect of the age of the cellar and the practice of inoculation. International Journal of Food Microbiology 103: 49–56.  Schirmer, M., U. Z. Ijaz, R. D’Amore, N. Hall, W. T. Sloan, and C. Quince. 2015. Insight into biases and sequencing errors for amplicon sequencing with the Illumina MiSeq platform. Nucleic Acids Research gku1341  Schlosser, J. W., A. G. Reynolds, M. King, and M.A. Cliff. 2005. Canadian terroir: Characterization of Chardonnay wines from the Niagara peninsula. Food Research International 38: 11-18.  Schoch, C.L., Seifert, K.A., Huhndorf, S., Robert, V., Spouge, J.L., Levesque, C.A., Chen, W., Bolchacova, E., Voigt, K., Crous, P.W., et al. 2012. Nuclear ribosomal internal transcribed spacer (ITS) region as a universal DNA barcode marker for fungi. Proceedings of the National Academy of Science 109: 6241-6246. 	 87	 Scholl, C.M., S. C. Morgan, M. L. Stone, M. Tantikachornkiat, M. Neuner, and D. M. Durall. 2016. Composition of Saccharomyces cerevisiae strains in spontaneous fermentations of Pinot noir and Chardonnay. Australian Journal of Grape and Wine Research. In press.   Schuster, SC. 2008. Next-generation sequencing transforms today’s biology. Nature Methods 5: 16-18.  Schutz, M., and J. Gafner. 1993. Analysis of yeast diversity during spontaneous and inoculated alcoholic fermentations. Journal of Applied Bacteriology 75: 551-558.  Styger, G., D. Jacobson, and F.F. Bauer. 2011. Identifying genes that impact on aroma profiles produced by Saccharomyces cerevisiae and the production of higher alcohols. Applied Microbiology and Biotechnology 91: 713–30.  Suárez-Lepe, J.A., and A. Morata. 2012. New trends in yeast selection for winemaking. Trends in Food Science and Technology 23: 39–50.  Sumby, K.M., P.R. Grbin and V. Jiranek. 2010. Microbial modulation of aromatic esters in wine: Current knowledge and future prospects. Food Chemistry 121:1-16.
  Swiegers, J.H., E.J. Bartowsky, P.A. Henschke, and I.S. Pretorius. 2005. Yeast and bacterial modulation of wine aroma and flavour. Australian Journal of Grape and Wine Research 11: 139–173.  Torija, M.J., N. Rozès, M. Poblet, J.M. Guillamón, and A. Mas. 2001. Yeast population dynamics in spontaneous fermentations: comparison between two different wine-producing areas over a period of three years. Antonie van Leeuwenhoek 79: 345–352.  Ugliano, M. and P.A. Henschke. 2009. Yeasts and Wine Flavour. Wine Chemistry and Biochemistry. Springer. New York. pp. 313-392. 	 88	 Varela, C., T. Siebert, D. Cozzolino, L. Rose, H. Mclean, and P. A. Henschke. 2009. Discovering a chemical basis for differentiating wines made by fermentation with “wild” indigenous and inoculated yeasts: role of yeast volatile compounds. Australian Journal of Grape and Wine Research 15: 238–248.  Vilanova, M., and C. Sieiro. 2006. Contribution by Saccharomyces cerevisiae yeast to fermentative flavour compounds in wines from cv. Albariño. Journal of Industrial Microbiology and Biotechnology 33: 929–933.  Zott, K., C. Miot-Sertier, O. Claisse, A. Lonvaud-Funel, and I. Masneuf-Pomarede. 2008. Dynamics and diversity of non-Saccharomyces yeasts during the early stages in winemaking. International Journal of Food Microbiology 125: 197-203.   Zott, K., C. Thibon, M. Bely, A. Lonvaud-Funel, D. Dubourdieu, and I. Masneuf-Pomarede. 2011. The grape must non-Saccharomyces microbial community: impact on volatile thiol release. International Journal of Food Microbiology 151: 210–215. 	 89	 Appendices Appendix A. Fermentation details during inoculated and spontaneous fermentations of Pinot noir and Chardonnay at Quails’ Gate Estate Winery Table A1. Inoculated Pinot noir fermentation  Fermentation Data of QGEW Pinot noir Inoculated Tanks Data Tank 1 Tank 2 Tank 3 Winery tank ID 13PNDRV-R20 13PNDRV-R21 13PNDRV-R22 Reference tank ID R20 R21 R22 Vintage (year) 2013 2013 2103 Fermentation treatment Inoculated Inoculated Inoculated Varietal L. var Pinot noir L. var Pinot noir L. var Pinot noir Field/block DRV DRV DRV Date of Harvest/crush 2013-10-10 2013-10-10 2013-10-10 SO2 addition 40ppm 40ppm 40ppm Cold soak Date of sample 2013-10-11 2013-10-11 2013-10-11 Depth of sample ~1m ~1m ~1m Temperature of tank (°C) 14 15 14 pH 3.41 3.37 3.34 Residual sugar (°Brix) 24 25 25 Cell density (CFU/mL) 1.42x10^7 2.81x10^7 3.02x10^7 Inoculation Commercial inoculum Lalvin RC212 Lalvin RC212 Lalvin RC212 Date of inoculation 2013-10-13 2013-10-13 2013-10-13 Cell density (CFU/mL) 1.77x10^7  Early Date of sample 2013-10-16 2013-10-16 2013-10-15 Depth of sample ~1m ~1m ~1m Temperature of tank (°C) 19 22 19 pH 3.46 3.44 3.67 Residual sugar (°Brix) 16 18 18 Cell density (CFU/mL) 1.62x10^7  1.14x10^7  2.07x10^7  Mid Date of sample 2013-10-17 2013-10-17 2013-10-16 Depth of sample ~1m ~1m ~1m Temperature of tank (°C) 26 27 22 pH 3.59 3.49 3.42 Residual sugar (°Brix) 5 3 14.5 Cell density (CFU/mL) 2.31x10^7  1.8x10^7  2.21x10^7 Late Date of sample 18.13.10 18.13.10 17.13.10 Depth of sample ~1m ~1m ~1m Temperature of tank (°C) 26 25 23 pH 3.59 3.62 3.49 Residual sugar (°Brix) -0.5 -0.4 0.5 Cell density (CFU/mL) 1.13x10^7  1.96x10^7  2.04x10^7  Final Notes Date pressed  2013-10-22 2013-10-22 2013-10-22 Duration of fermentation 11 days 11 days 11 days   	 90	Table A2. Spontaneous Pinot noir fermentation Fermentation Data of QGEW Pinot noir Spontaneous Tanks Data Tank 1 Tank 2 Tank 3 Winery tank ID 13PNQG-R10 13PNQG-R11 13PNQG-R12 Reference tank ID R10 R11 R12 Vintage (year) 2013 2013 2103 Fermentation treatment Spontaneous Spontaneous Spontaneous Varietal L. var Pinot noir L. var Pinot noir L. var Pinot noir Field/block F3B8/F4B7 F4B8/F4B9 F4B8/F4B9 Date of Harvest/crush 2013-09-27 2013-09-27 2013-09-27 SO2 addition 0ppm 0ppm 0ppm Cold soak Date of sample 2013-09-30 2013-09-30 2013-09-29 Depth of sample ~1m ~1m ~1m Temperature of tank (°C) 15 15 15 pH 3.35 3.1 3.18 Residual sugar (°Brix) 25 24.5 24 Cell density (CFU/mL) 3.7x10^6 CFU/mL 4.1x10^6 CFU/mL 5.4x10^6 CFU/mL Early Date of sample 2013-10-02 2013-10-02 2013-10-02 Depth of sample ~1m ~1m ~1m Temperature of tank (°C) >20 22 23 pH 3.41 3.41 3.42 Residual sugar (°Brix) 22 20 20 Cell density (CFU/mL) 2.9x10^7 CFU/mL 2.88x10^7 CFU/mL 2.76x10^7 CFU/mL Mid Date of sample 2013-10-03 2013-10-03 2013-10-03 Depth of sample ~1m ~1m ~1m Temperature of tank (°C) 26.5 26 27 pH 3.43 3.45 3.44 Residual sugar (°Brix) 12 11 11.5 Cell density (CFU/mL) 2.37x10^7  2.81x10^7  2.56x10^7  Late Date of sample 2013-10-07 2013-10-07 2013-10-07 Depth of sample ~1m ~1m ~1m Temperature of tank (°C) 23 23 24 pH 3.6 3.57 3.59 Residual sugar (°Brix) 1.8 1 1 Cell density (CFU/mL) 1.23x10^7  2.48x10^7  2.6x10^7  Final Notes Date pressed  2013-10-13 2013-10-13 2013-10-13 Duration of fermentation 13 days 13 days 13 days    	 91	Table A3. Inoculated Chardonnay fermentation Fermentation Data of QGEW Chardonnay Inoculated Barrels Data Barrel 1 Barrel 2 Barrel 3 Winery Barrel ID 13CHF5Upper11149 13CHF5Upper11061 13CHF5Upper11148 Reference Barrel ID 11148 11149 11061 Vintage (year) 2013 2013 2013 Fermentation treatment Inoculated Inoculated Inoculated Varietal Chardonnay Chardonnay Chardonnay Field/block 13CHF5Upper 13CHF5Upper 13CHF5Upper Date of Harvest/crush 2013-09-30 2013-09-30 2013-09-30 Cold soak- W43 Date of sample 2013-10-01 Depth of sample ~1m Temperature of Tank (°C) 10 pH 3.19 Residual sugar (°Brix) 22.5 Cell density (CFU/mL) 1.16x10^7 SO2 addition 40ppm Inoculation Commercial inoculum Zymaflore VL2 Date of inoculation 2013-10-02 Cell density (CFU/mL) 3.01x10^7 Early Date of sample 2013-10-06 2013-10-06 2013-10-06 Depth of sample ~12inches ~12inches ~12inches Temperature of Barrel (°C) 17.5 17.5 17.5 pH 3.06 3.06 3.06 Residual sugar (°Brix) 18.5 18.5 18.5 Cell density (CFU/mL) 9.7x10^6  1.35x10^7  8.8x10^6  Mid Date of sample 2013-10-10 2013-10-10 2013-10-10 Depth of sample ~12inches ~12inches ~12inches Temperature of Barrel (°C) 22 22 22 pH 3.18 3.18 3.18 Residual sugar (°Brix) 6 6 6 Cell density (CFU/mL) 1.11x10^7 7.0x10^6  1.32x10^7  Late Date of sample 2013-10-16 2013-10-16 2013-10-16 Depth of sample ~12inches ~12inches ~12inches Temperature of Barrel (°C) 18.5 18.5 18.5 pH 3.36 3.36 3.36 Residual sugar (°Brix) 3.5 3.5 3.5 Cell density (CFU/mL) 4.4x10^6  5.5x10^6  5.2x10^6  Final Notes Duration of fermentation 18 days 18 days 18 days   	 92	Table A4. Spontaneous Chardonnay fermentation Fermentation Data of QGEW Chardonnay Spontaneous Barrels Data Barrel 1 Barrel 2 Barrel 3 Winery Barrel ID 13CHF5Upper11013 13CHF5Upper11208 13CHF5Upper11209 Reference Barrel ID 11013 11208 11209 Vintage (year) 2013 2013 2013 Fermentation treatment Spontaneous Spontaneous Spontaneous Varietal Chardonnay Chardonnay Chardonnay Field/block 13CHF5Upper 13CHF5Upper 13CHF5Upper Date of Harvest/crush 2013-09-30 2013-09-30 2013-09-30 Cold soak- W43 Date of sample 2013-10-01 Depth of sample ~1m Temperature of Tank (°C) 10 pH 3.19 Residual sugar (°Brix) 22.5 Cell density (CFU/mL) 1.16x10^7 CFU/mL SO2 addition 40ppm Early Date of sample 2013-10-07 2013-10-07 2013-10-07 Depth of sample ~12inches ~12inches ~12inches Temperature of Barrel (°C) 16 16 16 pH 3.17 3.17 3.17 Residual sugar (°Brix) 19 19 19 Cell density (CFU/mL) 8.9x10^6  1.71x10^7  1.06x10^7  Mid Date of sample 2013-10-10 2013-10-10 2013-10-10 Depth of sample ~12inches ~12inches ~12inches Temperature of Barrel (°C) 20 20 20 pH 3.19 3.19 3.19 Residual sugar (°Brix) 6.5 6.5 6.5 Cell density (CFU/mL) 1.84x10^7  1.87x10^7  2.11x10^7  Late Date of sample 2013-10-17 2013-10-17 2013-10-17 Depth of sample ~12inches ~12inches ~12inches Temperature of Barrel (°C) 16 16 16 pH 3.3 3.3 3.3 Residual sugar (°Brix) 4 4 4 Cell density (CFU/mL) 4.9x10^6  1.08x10^7  8.7x10^6  Final Notes Duration of fermentation 18 days 18 days 18 days 	 93	 Appendix B. Culture dependent and independent identities of all yeasts present across inoculated and spontaneous Pinot noir and Chardonnay fermentations  Table B1. Culture dependent yeast isolate identities within Pinot noir     Inoculated Pinot noir (n=3) Cold soak Early Mid Late Yeast isolate Average relative abundance (%) Yeast isolate Average relative abundance (%) Yeast isolate Average relative abundance (%) Yeast isolate Average relative abundance (%) H. uvarum 98.6 Lalvin RC212 34.7 Lalvin RC212 52.8 Lalvin RC212 44.4 P. kluyveri 1.4 Lalvin D254 22.2 Vitilevure 3001 26.4 P. occidentalis 36.1   Vitilevure 3001 20.8 P. occidentalis 15.3 Vitilevure 3001 8.3   P. occidentalis 13.9 H.uvarum 4.2 Redstart Montrachet 4.2   Fermol Arome plus 2.8 Fermol Arome plus 1.4 H. uvarum 4.2   Redstar Montrachet 2.8   C. diversa 1.4   Lallemand BM45 1.4   C. albicans 1.4   C. californica 1.4     Spontaneous Pinot noir (n=3) Cold soak Early Mid Late Yeast isolate Average relative abundance (%) Yeast isolate Average relative abundance (%) Yeast isolate Average relative abundance (%) Yeast isolate Average relative abundance (%) H. uvarum 95.8 Lalvin D254 47.2 Lalvin D254 45.8 Lalvin D254 56.9 H. meyeri 1.4 Vitilevure 58W3 15.3 Lalvin RC212 18.0 Vitilevure 58W3 12.5 H. opuntiae 1.4 Fermol Arome plus 11.1 Vitilevure 58W3 11.1 Unknown 1 6.9 C. diversa 1.4 Lalvin RC212 7.0 Rhone 4600 4.2 Fermol Arome plus 5.5   Unknown 1 4.2 Lalvin CY3079 4.2 Unknown 5 4.2   Zymaflore VL2 2.8 Unknown 1 4.2 Lallemand BM45 1.4   Unknown 2 2.8 Unknown 3 2.8 Rhone 4600 1.4   Unknown 5 2.8 Fermol Arome plus 1.4 Lalvin RC212 1.4   Unknown 8 2.8 Unknown 2 1.4 Zymaflore VL3 1.4   Fermol Blanc 1.4 Unknown 4 1.4 Lalvin CY3079 1.4   H. uvarum 1.4 Unknown 7 1.4 Unknown 2 1.4   Unknown 6 1.4 Unknown 9 1.4 Unknown 3 1.4     Unknown 10 1.4 Unknown 6 1.4     Unknown 11 1.4 Unknown 7 1.4       Unknown 8 1.4 	 94	 Table B2. Culture dependent yeast isolate identities within Chardonnay  Inoculated Chardonnay (n=3) Cold soak Early Mid Late Yeast isolate Average relative abundance (%) Yeast isolate Average relative abundance (%) Yeast isolate Average relative abundance (%) Yeast isolate Average relative abundance  (%) H. uvarum 100 Lalvin D254 47.2 Zymaflore VL2 52.8 Zymaflore VL2 54.2   Fermol Arome plus 15.3 P. occidentalis 47.2 P. occidentalis 34.7   Unknown 2 12.2   Lalvin RC212 8.3   Zymaflore VL2 9.8   H. uvarum 1.4   Lalvin CY3079 2.8   S. uvarum 1.4   Unknown 4 2.8       Unknown 9 2.8       Lalvin RC212 1.4       Rhone 4600 1.4       C. zemplinina 1.4       Unknown 1 1.4       Unknown 7 1.4     Spontaneous Chardonnay (n=3) Cold soak Early Mid Late Yeast isolate Average relative abundance (%) Yeast isolate Average relative abundance (%) Yeast isolate Average relative abundance (%) Yeast isolate Average relative abundance  (%) H. uvarum 100 Lalvin D254 38.9 P. occidentalis 72.2 Zymaflore VL2 62.6   Fermol Arome plus 18 Zymaflore VL2 23.6 P. occidentalis 37.4   Zymaflore VL2 8.3 Lalvin CY3079 2.8     Unknown 2 9.7 H. uvarum 1.4     Lalvin RC212 8.3       Unknown 3 4.2       Rhone 4600 2.8       Lalvin CY3079 2.8       Unknown 4 2.8       Unknown 1 1.4       Unknown 5 1.4       Unknown 12 1.4      	 95	 Table B3. Culture independent dominant yeast species identities within Pinot noir   Inoculated Tanks (n=3) Cold soak Early Mid Late Yeast species Average relative abundance (%) Yeast species Average relative abundance (%) Yeast species Average relative abundance (%) Yeast species Average relative abundance (%) H. uvarum 48.25  S. cerevisiae 83.51  S. cerevisiae 87.98  S. cerevisiae 84.69  A. pullulans 16.22  H. uvarum 12.43  H. uvarum 6.72  H. uvarum 10.82  Ascomycota ssp 10.16  Pichia ssp 2.26  Pichia ssp 2.66 Pichia ssp 2.45  Saccharomycete ssp 8.65 Total (%) 98.2 Total (%) 97.4 Total (%) 97.9 Pichia ssp 2.13        D. tassiana 1.69        C. californica 1.11        K. aerobia 1.11       Total (%) 89.3       Spontaneous Tanks (n=3) Cold soak Early Mid Late Yeast species Average relative abundance (%) Yeast species Average relative abundance (%) Yeast species Average relative abundance (%) Yeast species Average relative abundance (%) H. uvarum 54.69  S. cerevisiae 69.79  S. cerevisiae 65.88  S. cerevisiae 69.58 Saccharomycete ssp 32.34  H. uvarum 27.08  H. uvarum 31.06  H. uvarum 26.77  R. nothofagi 2.79  Total (%) 96.9 Total (%) 96.9 Total (%) 96.4 A. pullulans 2.32        Ascomycota ssp 1.89        B. caroliniana 1.49        Pichia ssp 1.32        Total (%) 96.8        	 96	 Table B4. Culture independent dominant yeast species identities within Chardonnay  Cold Soak Tank (n=1) Yeast species Relative abundance (%) B caroliniana 29.39 H. uvarum 18.56 Saccharomycete ssp. 16.76 Ascomycota ssp 12.39 D. tassiana 5.89 S. cerevisiae 2.59 K. aerobia 2.19 M. bainieri 1.53 Total (%) 89.3 Inoculated Barrels (n=3) Early Mid Late Yeast species Average relative abundance (%) Yeast species Average relative abundance (%) Yeast species Average relative abundance (%) S. cerevisiae 91.12  S. cerevisiae 88.19  S. cerevisiae 86.17  Pichia ssp. 6.65  Pichia ssp. 10.50  Pichia ssp. 13.14  Total (%) 97.8 Total (%) 98.7 Total (%) 99.3 Spontaneous Barrels (n=3) Early Mid Late Yeast species Average relative abundance (%) Yeast species Average relative abundance (%) Yeast species Average relative abundance (%) S. cerevisiae 92.76 S. cerevisiae 90.63  S. cerevisiae 88.02  Pichia ssp. 4.83 Pichia ssp. 7.99  Pichia ssp. 10.89 Total (%) 97.6 Total (%) 98.6 Total (%) 98.9 	 97	Appendix C. Three-way ANOVAs for Pinot noir and Chardonnay sensory evaluations  Table C1. Pinot noir   F- ratio Attribute Judges Wines Replicates Judge x Wines Judge x Replicates Wines x Replicates Mean square error Cherry aroma 35.98*** 2.30** 1.27 1.87* 2.37* 0.7 91.51 Red berry aroma 29.29*** 2.14** 2.54 2.35** 2.09 1.16 100.6 Black berry aroma 37.40*** 0.74 0.99 1.62 4.42** 0.42 72.7 Jammy aroma 57.52*** 0.64 2.36 3.09** 1.83 1.63 58.97 Prune aroma 15.81*** 0.70 0.60 1.87* 7.43*** 1.02 130.7 Vegetative aroma 34.49*** 3.30** 5.90* 1.75 5.44*** 0.32 97.32 Dill aroma 24.75*** 1.47** 5.648 3.07** 4.91** 1.23 106.3 Black pepper aroma 52.68*** 4.65* 0.02 0.71 2.30* 0.18 105.1 Earthy aroma 12.46*** 2.19* 0.21 1.63 1.91 1.07 205.9 Cherry flavour 35.27*** 1.81** 14.71** 2.6** 8.43*** 0.58 86.26 Red berry flavour 44.07*** 0.81 10.32** 1.62 1.71 0.35 111.9 Black berry flavour 127.97*** 0.85* 12.16** 2.89** 3.54** 1.06 33.41 Jammy flavour 39.85*** 0.53 6.73* 1.41 8.19*** 2.37 46.85 Prune flavour 6.85*** 0.45 0.82 1.51 4.17** 1.00 110.4 Vegetative flavour 9.13*** 4.25*** 0.81 2.33** 5.91*** 0.48 121.3 Dill flavour 20.27*** 1.73* 2.87 2.07* 2.73* 0.99 130.9 Black pepper flavour 124.23*** 3.35*** 6.62* 2.10* 3.38** 1.12 54.12 Earthy flavour 11.10*** 4.45** 0.78 1.01 1.68 0.80 179.1 Sweetness 130.55*** 1.2* 24.35*** 2.57** 47.92*** 0.24 11.93 Acidity 22.77*** 1.41* 23.41*** 1.82* 9.64*** 1.31 58.15 Body/mouthfeel 12.99*** 0.96** 0.06 3.85*** 3.84** 0.43 86.05 Length of aftertaste 8.83*** 0.22 7.00* 2.05* 7.94*** 0.98 73.69 Colour intensity 16.13*** 94.32*** 13.79** 2.53** 1.25 0.88 35.40 Degrees of freedom 7 5 1 35 7 5 35 Significant at < 0.05*, <0.01**, <0.001***  	 98	 Table C2. Chardonnay   F- ratio Attribute Judges Wines Replicates Judge x Wines Judge x Replicates Wines x Replicates Mean square error Citrus aroma 20.46*** 0.94 1.97 1.01 2.93 1.00 139.5 Pome aroma 16.94*** 1.97 0.76 0.57 1.96 1.23 108.7 Tropical aroma 19.08*** 0.85 0.87 0.66 1.64 0.45 131.6 Peach aroma 20.03*** 1.57 0.75 0.81 0.51 0.92 126.3 Spice aroma 115.36*** 1.64 0.54 1.31 4.89 0.95 45.50 Oak/Vanilla aroma 55.65*** 1.23 34.66*** 4.42*** 2.48* 0.8 34.54 Honey aroma 51.78*** 0.69 8.82** 0.44 4.72** 0.16 83.57 Floral aroma 30.00*** 1.44 4.84* 0.89 1.57 0.54 93.49 Citrus flavour 115.89*** 0.53 8.33** 2.14* 15.23*** 0.92 38.75 Pome flavour 35.65*** 0.95 4.84* 0.87 5.22*** 0.61 86.95 Tropical flavour 76.08*** 0.99 6.94* 2.7** 3.5** 0.93 32.21 Peach flavour 31.27*** 0.52 18.55*** 0.99 5.73*** 0.89 93.68 Spice flavour 43.35*** 0.17 0.13 1.98* 1.93 0.69 40.88 Oak/Vanilla flavour 38.68*** 0.74 2.69 1.76* 8.06*** 1.14 46.89 Honey flavour 105.42*** 2.38 6.89* 0.68 2.21* 1.98 44.18 Floral flavour 72.47*** 1.87 0.00 1.01 2.43* 1.43 48.80 Sweetness 53.28*** 0.56 0.21 0.87 4.10** 1.16 61.45 Acidity 41.99*** 1.22 8.60** 0.97 0.55 0.97 51.43 Body/mouthfeel 26.95*** 0.98 12.44** 1.17 6.24*** 0.85 57.91 Length of aftertaste 93.91*** 0.31 4.29* 2.26** 12.29*** 2.24 27.80 Degrees of freedom 7 5 1 35 7 5 35 Significant at < 0.05*, <0.01**, <0.001***    

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0228805/manifest

Comment

Related Items