Harvest Guide 2019

Page 1

HARVEST 2019

H A RV E S T GUIDE 2019

PRESENTED BY ETS LABS


TA B L E O F C O N T E N T S

pages 04-05

pages 20-23

Maturity Monitoring

Scorpions

The ETS Grape Maturity Monitoring Panels provide set of analyses requested to monitor fruit maturity.

Find out what's coming in on your grapes with ScorpionsTM genetic detection.

Recognize the conditions that lead to VA formation and how to monitor the microbes that cause it.

pages 06-17

pages 24-25

pages 28-29

Phenolics

Potential Alcohol

Take a look at building your phenolics program and each step that is involved.

Get a closer look at using glucose+fructose analysis to estimate potential alcohol.

ETS Juice Panel

Get the complete picture for informed winemaking with the harvest Juice Panel

pages 18-19

2

pages 26-27

Volatile Acidity

pages 30-31

ETS Vintage Portal

Sugar Analysis

Learn all about our newest tool just in time for Harvest.

Sugar can mean a number of things. See what's behind your "Residual Sugar" numbers.


pages 32-33

pages 42-43

page 51

Aromas

Quercetin

Harvest Satellite Analysis

Detect and prevent common (and uncommon) sensory flaws.

Quercetin precipitates in bottled wines are particularly nasty as a clear, apparently stable wine at bottling can form a greenish goopy precipitate after bottling aging.

Use this quick reference to see which Juice Analyses are offered at your satellite lab.

pages 34-39

pages 44-49

pages 52-53

Smoke Impact

Harvest Toolkit

Sampling and Shipping

Take a look at the affects of smoke impact on berries and red wine.

This short guide will give you the highlights of our most requested Harvest testing.

Make sure you get the most out of your results using these sampling guidelines.

pages 40-41

page 50

pages 54-59

New Smoke Panels

New Near You

Our Locations

ETS is now offering an extended panel for volatile smoke markers and a glycosylated markers panel is coming this harvest!

Discover all the new resources we've put in place in time to make your harvest even simpler.

p. 55- St. Helena p. 56- Healdsburg p. 57- Paso Robles p. 58- Newberg p. 59- Walla Walla

3


MATURITY

MONITORING THE ETS GRAPE MATURITY MONITORING PANELS AND SUGAR PER BERRY PANELS

W H AT A R E T H E Y ? The ETS Grape Maturity Monitoring Panels provide set of analyses requested to monitor fruit maturity. They include the traditional measurements of juice solids (Brix) and acidity (Titratable Acidity and pH), enhanced by the addition of a more accurate determination of fermentable sugars (glucose + fructose) and malic acid, as the degradation of this organic acid is a well-known marker of ripening. They also include berry size parameters (volume and/ or weight), as well as a less familiar measurement: sugar per berry. In the ETS Newberg (OR) and Paso Robles (CA) satellite labs, the determination of sugar per berry is based on the sample’s average berry weight. In Walla-Walla (OR), Healdsburg (CA) and Saint Helena (CA) this calculation is based on the average berry volume as measured by Dyostem. The Dyostem also provides an assessment of berry volume variability expressed as a Coefficient of Variation (%) and a histogram of the berry volume distribution is provided in our reports. 4

When juice param et e rs a re not request ed, all be rr y param et ers descr ibed a bove are offered as our Sug a r p e r Ber r y panels: Brix, Berr y Weight and Sugar per Berr y (by weight) in Newberg (OR) and Paso Robles (CA) •

Brix, Berr y Weight, Berr y Volume, Berr y Volume Variability, and Sugar per Berr y (by volume) in Walla-Walla (OR), Healdsburg (CA) and Saint Helena (CA) •


W H Y M O N I TO R I N G S UGAR P ER B ERRY ? Monitoring Sugar per Berry, starting shortly after veraison, growers to determine the duration and rate of sugar loading. During this phase, vines synthesize and actively transport sugar into berries. At the end of this phase, Brix usually keeps increasing due to berry dehydration, however, and simply monitoring Brix cannot determine when the sugar loading phase stops. Conversely, when this point is reached, the actual quantity of sugar accumulated in each berry remains unchanged (see Fig 1).

SUGAR QUANTITY PER BERRY (MG)

FIG 1: THE SUGAR LOADING CONCEPT

A LOO K AT O U R DYOS TEM P C + TO U CH S CR EEN

ACTIVE SUGAR LOADING PERIOD

VERAISON

I N D U S TRIA L NUM E RIC CA MER A

NO MORE ACTIVE SUGAR LOADING

DAY «0 »

(PHYSIOLOGICAL RIPENESS)

L ED L I G H TS

A NA LYSIS P L AT E

The duration of the sugar loading phase, the time at which this phase ends, as well as the Brix achieved at that time, are all important indicators of wine growing conditions. Excessive or insufficient vigor, water availability and resistance to heat stress all have an impact. Vines under often desired moderate hydric stress conditions typically reach between 21 and 23 Brix at the end of sugar loading. In most cases vines with excessive vigor see their sugar loading stop before these levels are reached, as vegetative growth may compete with sugar synthesis and accumulation. Such vines also often poorly resist heat waves and tend to “shut down” more easily, which can prompt a sudden end of sugar accumulation in berries. At the other end of the spectrum, vines under excessive water stress, subject to nutrient deficiencies or diseased, also have difficulty reaching typical Brix levels at the end of sugar loading (see Fig 2). Furthermore, the end of sugar loading triggers a variety of maturity events influencing the development of grape aroma compounds and phenolics. It is often used as an indication of when to start grape phenolics measurements (see Vineyard Decisions – Grape Phenolic Panel p. 20).

POTE N TIAL D EG RE E AT SU GA R LOA D IN G STO P (% VO L. )

FIG 2: SUGAR LOADING & VINE WATER STATUS MERLOT 2012, PAUILLAC 14

13

MODE RAT E WAT E R RE ST RICT ION

12

NO WAT ER RES T R ICT ION

11

N UT RIT ION DE FICIE N CIE S (N , K )

WAT E R ST RE SS

H IG H Y IE LD 10

0

In Good Company ETS is proud to be par tnered with Fruition Sciences and Vivelys (Dyostem ®), to help you better characterize vineyards, closely monitor grape ripening, and make the most informed har vest decisions.

1 2 3 4 5 6 7 8 PR EDAWN LEAF WAT E R P OT E N T IA L DURIN G RI P E N I N G ( B A R, A B SOLUT E VA LUE ) 5


6


ETS

Juice Panel

W i n e m a ke r s re ly on j uice c he mis try a na lys is for a m ore com pl ete pi c t u re of mus t comp os ition at ha rve st th at goe s beyon d t ra ditiona l TA , pH, a nd °B rix. T his is critical a s ju i ce c h emis try ca n b e diffe re nt from vintage to v i nt a ge . Comb ining mode rn tool s g ive s vital i n s i ght s to ma ke informe d vineya rd ma na gem ent de c i si on s , c hoos e ha rve s t date s , p re dic t/adju st w i n e com p os ition a nd fa cil itate fe rme ntation s.

7


APPLYING JUICE CHEMISTRY TO WINEMAKING A thorough analytical picture gives winemakers the ability to better predict their wine composition, and plan appropriate winemaking strategies in response to changing must compositions. Whether you analyze your free-run juice, monitor your mid-fermentation chemistry, or do both, it is important to understand the analytical results in context within the fermentation process stage.

Analyzing juice pre-fermentation provides data that gives winemakers the opportunity to identify anything unusual about the current vintage, and to compare differences in composition from vintage to vintage. These insights affect winemaking decisions including acid adjustments and fermentation strategy.

E TS J UIC E PA NE L CO M PO N E N T S O B S ERV ED VARI ATI ON BE T W E E N V INTAG E S I N W H I T E G R AP E VARI ETALS

8

CA WHITE

MALIC

TARTARIC

Bri x

G LU/FRU

P O TAS S IUM

AM MON IA

NOPA

pH

TA

YAN

201 8

3.00

5.0

23. 1

240

147 2

67

12 5

3 .42

6.4

180

2 0 17

2.50

5.5

23.6

244

17 0 7

76

13 2

3 .52

5.9

194

2 0 16

2.50

4.9

23.6

247

153 4

72

158

3 .4 8

5.9

217

201 5

2.4 0

5.4

23.6

249

1553

84

13 1

3 .4 3

6.1

20 0

2 0 14

2.30

5.4

23.7

25 0

152 5

75

12 5

3 .4 6

5.8

187

OR WHITE

MALIC

TA RTA RIC

Bri x

G LU/FRU

P O TAS S IUM

AM MON IA

NOPA

pH

TA

YAN

2018

3.05

5.1

22. 1

228

1181

43

83

3 .2 5

7.0

119

2017

2.86

4.8

21 . 4

240

12 67

49

98

3 .2 6

6.8

135

2 0 16

2.17

5.3

22. 5

237

141 9

70

12 2

3 .3 8

5.9

180

201 5

3.03

5.3

22. 1

228

13 87

74

96

3 .3 0

6.8

131

2014

2.4 3

4.9

22. 9

233

13 57

60

88

3 .3 5

5.9

136

WA WHI TE

MALIC

TA RTA RIC

Bri x

G LU/FRU

P O TAS S IUM

AM MON IA

NOPA

pH

TA

YAN

2018

2.63

5.0

22. 4

230

12 51

44

93

3 .3 1

6.9

129

2017

2.86

5.1

21 . 9

228

13 98

56

86

3 .2 8

6.5

126

2 0 16

2.52

6 .0

22. 5

231

1514

66

13 4

3 .3 7

6 .6

187

201 5

2.21

4.8

23. 4

245

152 6

51

100

3 .47

5.3

142

2014

2.54

4.7

22.7

234

1498

61

88

3 .42

5.8

138


USING JUICE ANALYSIS TO PREDICT WINE COMPOS ITION Clients sometimes ask why they see a difference in the concentration of acids, or potential alcohol vs. final alcohol, when comparing their juice samples to mid- to end-stage fermentation samples. It is not unusual to observe differences between the levels of acids, potassium and sugar/potential alcohol in different samples from the same vineyard. Variations can occur depending on the vineyard sampling strategy and how representative the samples are of the vineyard. How juice samples are prepared matters, e.g. how thoroughly each grape sample is crushed and mixed. In white winemaking, the differences in composition between free run juices and the different press fractions are well known.

In red winemaking, free-run juices obtained after filling tanks may not reflect the actual content of the tank, since components such as acids and potassium can initially be sequestered at high levels in grape tissue next to the skin. As the grape tissue breaks down during cold soak, fermentation, and maceration, the resulting extraction of acids and potassium from the tissue into the juice can contribute to the observed differences. Likewise, the sugar in raisins or shriveled grapes may take a long time to release during red winemaking, causing an underestimate of fermentable sugar and therefore potential alcohol. Mid-point analysis on the fermenting wine (analyzing glucose+fructose and ethanol) may give a more accurate picture of fermentable sugar and potential alcohol. All of these factors can contribute to the differences observed between juice samples and the final wine composition.

E T S JUIC E PA N E L CO M PO N E N T S OB S ERV ED VARI ATI ON BET W E E N V IN TAG E S I N RED G RAP E VARI ETALS CA RED

MA L IC

TARTARIC

Bri x

G LU/ F RU

P O TAS S IUM

AM MON IA

NOPA

pH

TA

YAN

2018

2.34

4.3

25 . 2

2 60

1654

45

95

3 .64

5.0

131

2017

1.9 3

4.7

25 . 4

2 64

2 0 11

62

108

3 .7 3

4.8

15 3

2016

2.2 2

4.2

25 .6

2 69

184 5

66

13 5

3 .69

4.9

189

2015

2.2 3

4.5

25 . 9

276

194 3

65

108

3 .69

4.9

16 0

2014

1.9 7

4.4

26 .0

275

180 5

58

99

3 .74

4.4

14 5

OR RED

MA L IC

TA RTA RIC

Bri x

G LU/ F RU

P O TAS S IUM

AM MON IA

NOPA

pH

TA

YAN

2018

2.4 0

3. 9

24.6

2 56

13 3 9

34

78

3 .52

5 .2

10 6

2017

2.43

3. 5

23.0

240

160 3

40

94

3 .56

5.1

122

2016

2.21

4.0

24. 3

2 58

167 5

58

111

3 .61

4.8

15 7

2015

2.3 7

4.0

24. 8

2 65

184 8

52

84

3 .68

4.7

124

2014

1.98

3. 9

24. 3

2 54

182 5

43

72

3 .69

4.3

1 14

WA RED

MAL IC

TA RTA RIC

Bri x

G LU/ F RU

P O TAS S IUM

AM MON IA

NOPA

pH

TA

YAN

2018

2.6 7

4.3

24. 9

2 59

1547

27

70

3 .53

5 .8

92

2017

2. 32

3.7

2 4.7

2 58

17 3 3

31

72

3 .57

5.0

93

2016

2.25

4.2

25 . 4

2 69

180 2

43

101

3 .69

4.6

134

2015

2.24

4.0

25 . 8

272

190 9

39

76

3 .7 5

4.4

10 9

2014

2. 14

3. 8

25 . 5

2 68

17 15

34

66

3 .68

4.3

93 9


ANALYZING WINE DURING FERMENTATION Because the must components are in a state of flux from cold soak to post malolactic fermentation, many winemakers prefer to make incremental adjustments rather than rely on one initial or massive adjustment. Winemakers who are targeting a certain TA and pH or ethanol level, for instance, often check their wine chemistry again at the fermentation midpoint. These mid-point numbers are used to make ongoing and final fermentation adjustments, making it easier for winemakers to achieve their target values and providing a more controlled outcome.

SU GAR CONTENT °Br i x i s no t a tru e measure of fermentable suga r. Two j u ices w ith i d e n ti c a l ° B ri x ma y h a ve ve r y d i fferen t f in al alco h o l con cen tra ti o n s du e to va r yin g a m ou n ts o f fermen table s u gars . Sugar concentration increases rapidly in grapes as they mature. This increase is usually due to sugar movement from the leaves to the fruit. During the final stages of berry development, berry dehydration may also contribute significantly to the final sugar concentration. °Brix is a measure of soluble solids in juice and must. The soluble solids in grape juice are primarily sugars. Organic acids, however, have a significant impact on brix, especially with unripe grapes. °Brix is used as an estimate of sugar concentration and often as a predictor of potential alcohol, but is not a true measure of fermentable sugar. Two juices with identical °Brix may have very different final alcohol concentrations due to varying amounts of fermentable sugars. Observations over the last five vintages indicate that the amount of glucose + fructose per degree brix varies slightly from vintage to vintage. This can have a significant impact on potential ethanol predictions based solely on Brix. The sum of glucose + fructose measures the two main sugars present in juice that can be fermented by yeast. This analysis provides a sound basis for estimates of potential ethanol in the wine. This additional analysis is an important supplement to °Brix testing when final ethanol predictions are critical.

10


Glu+Fru (g/L): Degree Brix Rat io, White Varietals 1 0 .8 1 0 .7 1 0 .6 CA 2014

1 0 .5

CA 2015 1 0 .4

CA 2016 CA 2017

1 0 .3

CA 2018

10.2 10.1 10 PI NOT GRI S

CHARDO N N AY

SAUVIGNON B LA N C

R I E S LI N G

VIOGNIER

Glu+Fru (g/L): Degree Brix Rat io, Red Varietals 1 0 .9 10.8 10.7 10.6

CA 2014

10.5

CA 2015

10.4

CA 2016 CA 2017

10.3

CA 2018

10.2 10.1 10 CABERNET SAUVIGNON

M ERL OT

SYRAH

PINO T NO IR

PE T IT E VE RD O T

CA B E RNE T F RA NC

G RE NA CH E

MALBEC

Z INFA ND E L

It's important to note that in ripe fruit, glucose + fructose numbers often appear higher than the corresponding °Brix results. This is because °Brix is measured as a percentage by weight, meaning brix values are greatly influenced by the density of juice. Glucose + Fructose is measured as weight by volume and is independent of juice density. A must with 23.3 °Brix will not have 23.3% by volume fermentable sugar.

11


NITROGEN COMPOUNDS Sluggish and stuck fermentations, coupled with sulfide formation, have become increasingly common and are often associated with deficiencies of yeast assimilable nitrogen in the must. However, excessive concentrations of certain nitrogen compounds have been associated with microbial spoilage and other fermentation problems. Knowledge of nitrogen status is critical for effective fermentation management. Nitrogen compounds are essential macronutrients for yeast, and are required for cell growth, multiplication, and yeast activity. As with other juice chemistry components, YAN values fluctuate vintage to vintage due to changes in ammonia and/or NOPA.

YAN Differences in Red Varietals by Vintage 350 300 250 CA 2 0 1 4

200

CA 2 0 1 5 CA 2 0 1 6

150

CA 2 0 1 7 100

CA 2 0 1 8

50 0

CA B ER N ET S A U V IG N O N

MER LO T

SYRAH

PINOT NOIR

PETITE VERDOT

CABE RNE T G RE NA CH E F RANC

MALBEC

Z INFA ND E L

YAN Differences in White Varietals by Vintage 300 250 200

CA 2 0 1 4 CA 2 0 1 5

150

CA 2 0 1 6 CA 2 0 1 7

100

CA 2 0 1 8

50 0

P INOT GR IS

CHARDONNAY

SAUVI GNON BL ANC

R I E S LI N G

VIOGNIER

Yeast assimilable nitrogen includes both alpha amino nitrogen (NOPA) and ammonia. Analysis of only alpha amino nitrogen or only ammonia nitrogen does not provide an accurate indication of total nitrogen status for a given must. Ammonia and NOPA can change independently of each other resulting in different ratios of these two YAN components. 12


AMMONIA Ammonia is the form of nitrogen nutrition most easily assimilated by yeast. Wineries routinely supplement nitrogen deficient musts with diammonium phosphate at the start of or during fermentation to provide adequate nitrogen levels. Additional ammonia analysis and adjustments during fermentation may also be beneficial in minimizing the risk of stuck fermentations and sulfide formation. Ammonia results are expressed as mg NH3 per liter. These values may be expressed as nitrogen equivalents by multiplying NH3 results by 0.82.

Ammo n i a D i f f e r e n c e s i n White Var ietals by Vintage 120 100 80

CA 2 0 1 4 CA 2 0 1 5

60

CA 2 0 1 6 CA 2 0 1 7

40

CA 2 0 1 8

20 0 P INOT GRI S

CHARDONNAY

SAUV I G N O N BLA N C

R I E S LI N G

VIOGNIER

A m m o n i a D iffer ences in R ed Var ietals by Vintage 12 0 100

80

CA 2014 CA 2015

60

CA 2016 CA 2017

40

CA 2018

20

0

CABERNET SAUVIGNON

M ERL O T

SYRA H

PINO T NO IR

PE T IT E VE RD O T

CA B E RNE T F RA NC

G RE NA CH E

MALBEC

Z IN FAN DE L

13


NOPA Alpha amino nitrogen, otherwise referred to as “Nitrogen by OPA”, or NOPA, is determined using a method specific for alpha amino groups. It is a measurement of primary amino acids usable by yeast. NOPA does not include proline, which is not utilized by yeast, or ammonia. NOPA results are expressed as mg nitrogen per liter.

NOPA Differences in White Varietals by Vintage 250 200 CA 2 0 1 4

150

CA 2 0 1 5 CA 2 0 1 6

100

CA 2 0 1 7 CA 2 0 1 8

50 0

P INOT GR IS CHARDONNAY SAUVI GNON BL ANC

RI E S LI N G

VIOGNIER

NOPA Differences in Red Varietals by Vintage 250

200

CA 2 0 1 4

150

CA 2 0 1 5 CA 2 0 1 6

100

CA 2 0 1 7 CA 2 0 1 8

50

0 CABERNET SAUVIGNON

14

MERLOT

SYRAH

PINOT NOIR

PETITE VERDOT

CABERNET FRANC

GRENACHE

MALBEC

ZINFANDEL


ACID BALANCE The a cid co mposition of m us t i s a com plex balance of free hydrogen ions, a cids, ac i d s alt s , and c at i ons . Concent rat i ons of these va rio us component s and t h ei r i nt erac t i ons i nfluence ma ny winema k ing p aram et ers . Vi nt age var i at i ons i n any of the co mponents t h at i m pac t ac i d balance c an res ult i n unexpected cha ng es i n t h e fi nal pH and TA of a wi ne. The principa l o bjecti ve of ac i d m anagem ent i s t o ac h i eve a nd ma inta in a pH fa vorable t o opt i m um wi ne balance and s t abi li t y.

TARTARIC ACID Tartaric acid is one of the two major organic acids found in grapes. It accumulates in grape tissue early during development and declines during ripening due to berry growth and dilution. Tartaric acid is not usually metabolized in grapes. It is present in grapes, must, and wine as a free acid and weak acid-salt complex. Tartaric acid-salts may precipitate, primarily as potassium bitartrate and calcium tartrate.

Both the formation and solubility of salts are affected by a balance of components that are in flux throughout the early life of a wine. An increase in the ratio of the free tartaric acid to the tartaric acid salts will cause a decrease in pH. This will affect the flavor, balance, and stability of the final product. Tartaric acid is commonly used to adjust the acid balance of juices and wines. Understanding tartrate interactions is important in designing appropriate acidification strategies.

Tar tar ic Differences in W hite Varietals by Vintage

Ta rt a ri c D i f f e re n c e s i n R e d Va ri e t a l s b y Vi n t a g e

8. 0

6.0

7. 0

5.0

6. 0 CA 2 0 1 4

5. 0

CA 2 0 1 5

4. 0

CA 2 0 1 6

3. 0

CA 2 0 1 7 CA 2 0 1 8

2. 0

CA 2014 CA 2015

3.0

CA 2016 CA 2017

2.0

CA 2018

1.0

1. 0 0. 0

4.0

PINOT GRIS

CHARDONNAY

SAUVIGNON BLANC

RIESLING

VIOGNIER

0.0

C ABER NET M ER LOT S AUV IGNON

S Y R AH

P INOT NOIR

P ETITE C ABER NET GR ENAC HE M ALBEC ZINFA ND EL V ER DOT FR ANC

15


MALIC ACID Malic acid accumulates early in berry development and declines during ripening due to dilution and respiration. Viticultural practices and grape cluster environments may directly affect respiration rates of malic acid. Malic acid levels affect pH and titratable acidity.

PH pH is a measure of free hydrogen ions in solution (which corresponds to the chemical definition of acidity) and is used as a gauge of wine acidity. Wine color, potassium bitartrate stability (cold stability), calcium stability, and molecular SO2 level are directly related to wine pH. pH is also critical in relationship to microbial stability, interactions of phenolic compounds, and color expression. Overall, the 2018 vintage had lower pH as compared to the 2017 vintage.

pH D i f f e r e n c e s i n White Varietals by Vintage 3. 70

Malic acid is converted to lactic acid during malolactic fermentation, causing the loss of an acid group. The effect of this acid reduction on pH depends upon the initial amount of malic acid and buffer capacity of the wine. Malolactic fermentation in wines containing low levels of malic acid and high buffer capacity will have little impact on wine pH. Malolactic conversion in wines with high malic acid and low buffer capacity can result in a substantial pH increase. Malic acid tends to be higher in cooler vintages. That said, 2018 had the highest malic acid levels for all varietals except Syrah as compared to the previous four vintages.

M al ic Differ ences in White Var ieta ls b y Vin ta g e 4. 00

3. 60

3. 50

3. 50

3. 00 CA 2014

3. 40

CA 2015

3. 30

CA 2016

3. 20 3. 10

1. 50

CA 2018

1. 00

CA 2016 CA 2017 CA 2018

0. 50 0. 00 P I NO T G RIS

C H A R D O N N AY

SAUVIGNON BL ANC

RIESL ING

VIOGNIER

pH D i f f e r e n c e s i n R ed Varietals by Vintage

3. 50

3. 9

3. 00

3. 8

2. 50

3. 7

CA 2014 CA 2015

3. 6 3. 5 3. 4

PINO T G RIS

CH A RD O NNAY

SA U VIG NO N B L A NC

RIE SL ING

V I OGN I E R

M al ic Differ ences in Red Var ieta ls b y Vin ta g e

4

CA 2014

2. 00

CA 2016 CA 2017

1. 50

CA 2018

1. 00

CA 2015 CA 2016 CA 2017 CA 2018

0. 50

3. 3

0. 00

3. 2 CA B E RN E T S A U VI G N O N

16

CA 2015

2. 00

CA 2017

3. 00 2. 90

CA 2014

2. 50

MERLOT

SYRAH

P INO T NO IR

PETITE CABERN ET GREN ACHE MAL BEC VERDOT F RAN C

ZIN FANDEL

CABERNET SAUVI GNON

M ERLOT

SYRAH

PI NOT NOI R

PETI TE CABERNET GRENACHE MALBE C VERDOT FRANC

Z I N FAN DE L


POTASSIUM Potassium is the primary cation present in grape tissue. Potassium concentration in the berry is a function of root uptake and translocation. Both are strongly affected by viticultural factors including choice of rootstock, potassium fertilization, and canopy management.

TITRATABLE ACIDITY Titratable acidity (TA) measures total available hydrogen ions in solution. This measurement includes both the free hydrogen ions and the undissociated hydrogen ions from acids that can be neutralized by sodium hydroxide. TA is the most widely used measurement of acidity in wine. Although generally considered a simple parameter, titratable acidity is actually a reflection of complex interactions between the hydrogen ions, organic acids, organic acid-salts, and cations in solution. Often there is no direct correlation between TA and pH. Two musts with similar titratable acidity may have very different pH values.

Potassium moves into cells in exchange for hydrogen ions from organic acids. Potassium concentration is highest near the grape skin. Crushing, skin contact, and pressing all influence potassium levels. Potassium is a critical factor in pH, acid salt formation, tartrate precipitation, and buffer capacity. Decreases in juice potassium are associated with decreased juice pH. For example, 2018 vintage California Cabernet Sauvignon harvested in November, had approximately 22% lower potassium values as compared to 2017. The pH for the 2018 vintage averaged 3.78 versus 3.95 for the 2017 vintage.

Potas s ium Differ ences in W h ite Var ietals by Vintage 2000 1800 1600

CA 2016

1000

CA 2017

600

8.0

200 0

6.0 CA 2 0 1 4 CA 2 0 1 5

4.0

CA 2018

800 400

5.0

CA 2015

1200

TA D i f f e r e n c e s i n White Varietals by Vintage 7.0

CA 2014

1400

3000

PINO T G RIS

CH A RD O NNAY

SA U VIG NO N B L A NC

RIE SL ING

V I OGN I E R

Potas s ium Differ ences in Re d Var ietals by Vintage

CA 2 0 1 6

3.0

CA 2 0 1 7

2500

CA 2 0 1 8

2.0

2000

1.0 0.0

P I NO T G RIS

C H A R D O N N AY

SAUVIGNON BL ANC

RIESL ING

5.0 CA 2 0 1 4

4.0

CA 2 0 1 5

3.0

CA 2 0 1 6

2.0

CA 2 0 1 8

CA 2 0 1 7

1.0 P IN O T N O IR

PETITE VERDOT

CABERN ET GREN ACHE MAL BEC F RAN C

ZIN FAN DEL

CA 2017 CA 2018

500 0

6.0

SYRAH

CA 2016

1000

7.0

CA B E RN E T M E R L O T SA U VI G N O N

CA 2015

VIOGNIER

TA D i f f e r e n c e s i n R ed Varietals by Vintage

0.0

CA 2014

1500

CABERNET M ERLOT SAUVI GNON

SYRAH

PI NOT NOI R

PETI TE VERDOT

CABERNET GRENACHE MALBE C FRANC

Z I N FAN DE L

CONCLUSIONS Juice chemistry is the foundation for the resulting wine. Producing wines with specific targets for ethanol and acid balance, while avoiding fermentation-related problems, requires a thorough understanding of the juice chemistry. Individual components of the juice can and do change from vintage to vintage. Winemakers aware of these changes will be able to adjust their winemaking process, where necessary, to achieve their targeted wine styles.

17


ETS VINTAGE PORTAL Winemakers often say they remember the last vintage, and their best and worst vintages. For the last seven years ETS Laboratories has provided Post-Harvest seminars detailing the differences observed in juice chemistry panel components each harvest as compared to previous harvests. Each vintage exhibits differences in juice chemistry with trends that are similar in California, Oregon and Washington. While seminar attendees usually find something interesting in our presentations, many continued to suggest that making real-time available data would be a very useful tool. The ETS Vintage portal provides a new way to identify vintage trends in juice chemistry and grape phenolic development in real time. Discussions with clients suggest the ability to compare their juice panel and grape phenolic data with that of others would provide a tool to recognize vintage trends. In addition, it would enable them to see if their individual results are similar to others in their region growing a particular varietal. Early identification of trends in a vintage can help with picking and process decisions. The following graphs highlight some of the interesting differences observed between the 2017 and 2018 vintages.

18


Com paris o n b et ween 2 0 1 7 a n d 2018 YA N i n C a l i forn ia C abernet Sa uvign o n s h ow in g ove ra l l l owe r YA N va l ue s 300

250

200

2017 YAN 2018 YAN

150

2017 YAN 2018 YAN

100

50

0 8/13

8/23

9/2

9/12

9/22

10/2

10/12

10/22

11/1

11/11

Com paris o n B et ween 2017 a n d 2018 B r i x i n C a l i forn ia C abernet Sau vignon show ing d elayed B r ix a cc um ul a t i on wi t h h i g h e r f i nal Brix levels 32 30 28 2017 Brix

26

2018 Brix 2017 Brix

24

2018 Brix 22 20 18 8/13

8/23

9/2

9/12

9/22

10/2

10/12

10/22

11/1

JOIN US

11/11

C o m paris o n B et ween 2017 a n d 2018 Pot a s s i um i n C alifornia C abern et Sauvign o n s h ow in g la rg e de c re a s e s i n pot a s s i um l evels at har vest 3500 3000 2500

2017 Potassium 2018 Potassium

2000

2017 Potassium 2018 Potassium

1500 1000 500 8/13

8/23

9/2

9/12

9/22

10/2

10/12

10/22

11/1

11/11

Com paris o n B et ween 2017 a n d 2018 pH i n Ca l i fornia C abern et Sau vign on show ing lower p H val ue s a t h a r ve s t 4.5 4.3 4.1 3.9 2017 pH 2018 pH

3.7

2017 pH 2018 pH

3.5

We invite our clients to join the Vintage Portal community this harvest. The results from the Juice Panels and Grape Phenolic Panels, with no identifying information other than varietal and AVA, are going to be analyzed and posted daily. The only information we need is the varietal and AVA code for each sample. The ETS harvest labels for Juice and Berry samples have boxes to enter the two letter codes for the AVA and varietal. These codes can be found on the ETS website and at the front counter at each of the ETS locations. We look forward to working with you to provide this new and exciting resource to our clients.

3.3

JUICE & BERRY SAMPLES

3.1 2.9 8/13

Client #:

8/23

9/2

9/12

9/22

10/2

10/12

10/22

11/1

Client:

11/11 Sample ID:

Com paris o n B et ween 2 0 1 7 a n d 2018 t a n n i n Ca l i forn ia C abernet Sa uvign o n s h ow in g n o i n c re a s e dur i n g r i pe n i n g

 Juice Panel

1200 1000

 Malic Acid  NOPA

 TA  pH

 Ammonium  Potassium

 Combined  Bacteria

 Tartaric  Other:

800

2017 tannin 2018 tannin

600

2017 tannin 400

2018 tannin

N

V

C

 Ethanol

 TSO 2

 VA

 Grape Water Content

VAR

AVA

 Yeast

 FSO2

S

 Grape Maturity Panel  Grape Phenolic Panel

Enter the AVA and Varietal on your juice label. For more information go to help.etslabs.com and search for Vintage Portal.

200 0

Juice Scorpions™

 GLU+FRU  BRIX

8/8

8/18

8/28

9/7

9/17

9/27

10/7

10/17

10/27

11/6

19


JUICE SCORPIONS I n d i ge n ous mic robes coming into the win er y on fr ui t i s one o f t h e mos t imp o r ta nt entr y ro utes for sp oi lage organi s m s t ha t c an c au s e s tuck a nd sluggish ferment at i ons and VA p ro bl e ms . Identifying and quantifying yeast and bacteria that can cause spoilage during the winemaking process is the first step in preventing these spoilage problems. Using Scorpions to see the full picture of spoilage microbes in the juice from incoming fruit gives the winemaker better situational awareness for potential problems that can occur during the cold soak process, in stuck or sluggish fermentations, or later during wine aging.

Volatile acidity in juice Volatile acidity, measured as acetic acid, can be formed throughout the winemaking process. Both acetic acid bacteria and strains of wild yeast – particularly Hanseniaspora and Pichia – are commonly linked to volatile acidity production prior to and in the early stages of fermentation. Elevated VA levels often occur during the cold soak process, or between cold soak and fermentation during red wine production. The VA-producing spoilage microorganisms grow quickly during this time, producing increasing levels of acetic acid until fermentation conditions inhibit their growth. Production of high levels of volatile acidity prior to fermentation can also cause problems later in the production

20

process, including possible impacts on the fermentation performance and wine sensory attributes. Large numbers of Acetic Acid bacteria on incoming fruit can carry through the fermentation and cause problems with VA production when exposed to air during barrel aging.

Effects on fermentation In addition to causing sensory impacts, large populations of wild yeast can deplete the YAN in the must, resulting in a YAN deficiency for the Saccharomyces cerevisiae driving the fermentation. Winemakers who detect high levels of Hanseniaspora or Pichia in a must usually recheck YAN before yeast inoculation, and supplement YAN if necessary. Likewise, if the Scorpions assay detects heterofermentative lactic acid bacteria, such as Lactobacillus brevis, L. kunkeei, L. hilgardii, L. fermentum, and Oenococcus oeni, in a juice, winemakers usually increase their monitoring of malic acid and microbe levels if the fermentation becomes sluggish or stuck. Early identification of the presence of these bacteria and recognizing the risk they pose to difficult fermentations is key to preventing VA formation in stuck fermentations.


W H AT W E ' R E LOOKING FOR... Acetic acid bacteria Acetic acid bacteria are commonly associated with grapes and the winery environment. The three groups of commonly detected acetic acid bacteria are Gluconobacter, Gluconacetobacter and Acetobacter. Both Gluconacetobacter and Acetobacter can generate acetic acid from ethanol in the presence of oxygen. The presence of these organisms can cause elevated volatile acidity in wines exposed to air.

Hanseniaspora Hanseniaspora (Kloeckera) is a wild apiculate yeast that is often present at high levels on incoming fruit. Hanseniaspora can initiate fermentation in the must and produce high levels of volatile acids, including acetic acid and ethyl acetate. It has been associated with acid rot in grapes infected by Botrytis cinerea. Population levels usually decline as alcohol concentration increases.

Pichia Pichia is a wild yeast that is often present at high levels on incoming fruit. Pichia can initiate fermentation, resulting in production of high levels of volatile acids, including acetic acid and ethyl acetate. These yeast have been associated with films formed in barrels and tanks during storage.

21


THE MISSING LINK Juice chemistry and microbiology are closely linked – looking at both gives winemakers deeper insights into the final wine and can help highlight problems before they get out of control.

!

CASE STUDY: • A juice sample was submitted for standard juice panel analysis. The juice panel showed <0.05 g/L of malic acid • The wine experienced VA formation early in fermentation

ANALYSIS ON THE JUICE: br i x

23 .2º

g l uc o se+fruc to s e pH

221 g /L 3.3

t a r ta r ic a c id

5 . 21 g /L

L-m a l ic a c id

< 0.05 g /L

p ota ssium t itr ata bl e a c id it y

1 230 mg /L 4. 31 g /L

L acto b acil l u s brev is g roup L acto b acil l u s ku nke e i L acto b acil l u s cas e i g roup

Pe d io co ccu s s pp

<10 cells /mL

Ace t ic a cid b a cte ria

9,68 0 ce lls /mL

Bre ttanomyce s bru xe l l e ns is

<10 cells /mL 8 70 cells /mL

Zyg o s accharomyce s b ail ii

a mm on i a

1 10 mg /L

Hans e nias pora s pp Pichia s pp

RESULTS After running Scorpions analysis, it became clear that the juice had high levels of Lactobacillus kunkeei, which can cause spontaneous malolactic fermentation (resulting in loss of malic acid and formation of volatile acidity from sugar metabolism.) Combining chemistry and microbiology data earlier in the winemaking process would have allowed for early discovery of the bacteria load and prevention steps. 22

1,200 cells /mL <10 cells /mL

16 0 mg /L

25 1 m g/ L (as N )

>10,000,000 cells /mL

L acto b acil l u s pl antaru m

a l ph a -a m i no N ( NOPA)

YAN

10 cells /mL

12,8 00 ce lls /mL 9,68 0 ce lls /mL

Preventive Actions: • Monitor bacteria levels on incoming juice • If bacteria levels are elevated, protect juice from VA formation • Prioritize completion of primary fermentation If Necessary: • Physical removal of VA forming bacteria • Inactivation of VA forming bacteria


Z YG O S AC C H A R O M YC E S

Many wineries around the world use grape concentrate in the production of their wines. Although grape concentrate is not the only source of Zygosaccharomyces yeast, it is certainly a common source for Zygosaccharomyces to be introduced into the winery.

The primary species of Zygosaccharomyces typically associated with grape concentrate are Zygosaccharomyces bailii and Zygosaccharomyces bisporus. Recently, a different species, Zygosaccharomyces rouxii, was isolated from a fermenting concentrate sample provided by a client. In response to this finding, we modified the design of the Zygosaccharomyces Scorpions primer/ probe combination to detect more species. The new Zygosaccharomyces Scorpions diagnostic detects an additional 7 species of Zygosaccharomyces for a total of 9 species.

W H AT W E ' R E LOOKING FOR... Zygosaccharomyces bailii Zygosaccharomyces bisporus Zygosaccharomyces rouxii Zygosaccharomyces lentus Zygosaccharomyces mellis Zygosaccharomyces kombuchaensis Zygosaccharomyces parabailii Zygosaccharomyces pseudobailii Zygosaccharomyces pseudorouxii

23


BUILDI N G A PHE N O L IC S PRO G R A M ETS offers a full suite of advanced HPLC-based analytical tools to evaluate phenolic compounds in grapes, juice, fermenting must and wine. The range of phenolic analyses allows flexible use and implementation to suit individual needs.

VIN E YA R D D ECIS I ON S GRAPE PHENOLIC PANEL

GRAPE PHENOLIC PANEL

Phenolic compounds in red wine grapes are directly linked to eventual wine flavor, color and aging characteristics.

Successful winemaking strategies require accurate information on grape composition. Winemakers use this panel during fermentation to reach target levels of tannin for specific wine styles, to monitor seed extraction, for adjusting tannin modification through oxygenation and for decisions on extended maceration and pressing.

The grape phenolic panel can characterize site to site variation as well as within site differences. It works well as a prediction tool for describing vintage effects on potential phenolics and is a great tool for vineyard research projects. It is particularly sensitive to grape maturation. The changes in grape tannin are particularly important for red wine picking decisions. The grape phenolic panel can track changes in seed ripening, skin tannin extractability and tannin modification. Dilution and concentration effects on tannin and other phenolic components can be monitored with the Grape Phenolic Panel particularly when used in conjunction with Grape Water Content.

24

U ND ERSTA NDING RAW MATERIA LS

M A N AG I N G FERM EN TATI O N RAPID PHENOLIC PANEL FOR WINE Phenolic compounds are extracted from grapes during fermentation and maceration. Monitoring the phenolic composition of the must during fermentation can greatly enhance a winemaker’s control of the process. Juice bleeds, fermentation temperatures, pump-over or punch down regimes, the use of rack-and-return, oxygen or air additions and press timings can all be fine-tuned with feedback on changes in phenolic composition. With this information, winemakers can adapt winemaking practices to fit the vintage and successfully create wines of a target style.

S ETTI N G TARGE T S RAPID PHENOLIC PANEL Phenolic composition is one of the main components of red wine style. The amount of tannin and its composition is the foundation of a wine’s structure. There are no “correct” values for these parameters. A winery must define their own style as a brand and for individual products within that brand. Integrating phenolic information into stylistic choices requires an understanding of the impact of tannin on the sensory profile of wines. For a winery new to this information a good strategy is to analyze a selection of recent products. Recent production lots, finished wines and competitor’s products are good choices. Tasting products with analytical information allows winery staff to build the connections between taste and analytical information. Taste is the final arbiter of style but a clear understanding of the relationship of taste and analysis is needed to turn analytical information into action.


W I N E LOT C HA RACTER IZ AT ION RAPID PHENOLIC PANEL After the completion of fermentation/ maceration, a wine lot typically represents a specific vineyard and fermentation tank. This is an excellent point for collecting quality control data. A comprehensive review of production lots is a powerful tool for monitoring block to block variation and the effects of winemaking practices. Analysis of finished production lots early in the vintage is very useful for changing fermentation practices and targets later in the vintage.

BLEND ING RAPID PHENOLIC PANEL FOR WINE OR RED WINE PHENOLIC PROFILE Winemakers interested in consistent tannin and color levels benefit by comparing the phenolic profiles of bulk wines prior to blending. Potential blends can be compared to target phenolic levels and benchmarks prior to final blend preparation.

B OTTLED W I N E C HA RAC TERI ZATI O N RED WINE PHENOLIC PROFILE Many wineries establish QC benchmarks for phenolic content immediately after bottling. This is especially useful for determining product consistency and for monitoring wine development during aging.

FI N I S HE D W I N E EVA LUAT I ON S RED WINE PHENOLIC PROFILE A historical review of products from within a winery and evaluation of similar products from other producers is an excellent way to establish phenolic benchmarks. This is often the first step in building an integrated program of phenolic analyses. A careful review of finished wines combined with sensory evaluation and market feedback can identify program strengths and weaknesses. The identification of desirable levels for key phenolic components creates targets that can be incorporated into process control points in the vineyard and winery.

25


VOLATILE ACIDITY

Some might say that it’s a wine’s destiny to become vinegar. Wine containing elevated levels of acetic acid bacteria and exposed to oxygen will naturally produce acetic acid, the key component of vinegar. Acetic acid is synonymous with volatile acidity (VA) in wine. Although excessive VA production by microbes is a natural process, it is also an entirely preventable problem. By recognizing the conditions that lead to VA formation and monitoring the microbes that cause it, winemakers can act to prevent problems before they occur.

26


The two components commonly associated with "VA taints", Acetic Acid and Ethyl Acetate, can be formed by both yeast and bacteria - and in the case of bacteria, can be formed with or without oxygen present.

V O L AT I L E A C I D I T Y ( VA )

E T H Y L A C E TAT E

is strictly speaking a measure of the volatile acids in wine, although in the real world the contribution of volatile acids other than acetic acid is negligible. VA is a normal component of wines at moderate levels (normal concentrations range from 0.3-0.9 g/L of acetic acid), but very quickly becomes undesirable as levels rise. The sensory threshold is around 0.91.0 g/L, depending on the style of wine, and the U.S. government sets legal limits of 1.2 g/L in white wine and 1.4 g/L in red wine.

Although not an acid, ethyl acetate is considered by some to be a component of VA. As an ester formed by ethanol and acetic acid, it is often linked to increased production of VA. From a sensory point of view, ethyl acetate is often classified as a “VA taint”, and its "nail polish remover” odor is often a telltale sign of high VA. Similar to acetic acid, there is a fine line between complexity and spoilage: small amounts of ethyl acetate can contribute “fruitiness/sweetness” or other positive characteristics to a wine at low levels. Normal concentrations are usually less than 100 mg/L, while the sensory threshold is generally 130150 mg/L, depending on the wine style.

COLD SOAK / EARLY STAGE FERMENTATION Acetic Acid can be produced prior to fermentation by Acetic Acid Bacteria and wild yeast in compromised fruit. It’s unusual to see alcohol present in juice before fermentation, but if clusters experience fungal rot or other types of damage (such as bird or insect damage), wild yeast in the vineyard can begin fermenting the juice that is leaked out. Acetic Acid bacteria can then convert the alcohol to acetic acid, causing “sour rot” and leading to high VA levels before fermentation has even begun.

Ethyl Acetate is often produced in the early stages of fermentation, and can be a particular problem in native fermentations with a slow start. Native yeast, especially Hanseniaspora, are the main Ethyl Acetate producers at this stage. Note that the Hanseniaspora can consume most or all of the YAN in the must very early in the fermentation, although they will only produce alcohol up to around 6%. High levels of Hanseniaspora and low YAN concentrations can contribute to stuck fermentations.

PRIMARY FERMENTATION Acetic acid production in primary fermentation is generally caused by yeast, including Saccharomyces and other species, but can also be formed by bacteria. The native yeasts Hanseniaspora and Pichia can drive fermentations up until around 6-7% alcohol, at which point they become stressed by the alcohol. Saccharomyces is more competitive as it is tolerant to and produces higher alcohol levels. In certain situations, the native yeasts respond to the changing fermentation conditions by producing elevated levels of acetic acid.

Ethyl Acetate production during fermentation is significantly impacted by the yeast strain and fermentation temperature. Although Saccharomyces cerevisiae will produce ethyl acetate, research has indicated that some of the Saccharomyces bayanus strains are more likely to form ethyl acetate in cold fermentations.

Bacteria, usually Lactobacillus, can also generate acetic acid from sugar and can often produce high levels of VA in stuck and sluggish fermentations. Oenococcus oeni, the bacteria used for inoculating most malolactic fermentations can also produce acetic acid from fructose. This frequently occurs at the end of malolactic fermentation if there is still fermentable sugar remaining in the wine. 27


POTENTIAL ALCOHOL The amount of fermentable sugar (glucose and fructose) in juice and the average conversion rate of sugar into alcohol can be used to predict the potential alcohol level in wines.

28


Q : H O W A C C U R AT E A R E P O T E N T I A L A L C O H O L E S T I M AT E S ? A:

Our clients have reported that glucose + fructose values improve the quality of their predictions, but it is important to remember that yeast populations and fermentation conditions vary, and any prediction of potential alcohol is only an approximation. Alcohol conversion ratios can be variable, so it is possible your actual alcohol may be lower or higher than the estimate. Many of our clients have found that the conversion rates observed for their own yeasts and fermentation conditions remain relatively constant, and they use their internally observed conversion rates to calculate potential alcohol content based on their glucose + fructose values. With white wines, predictions are usually fairly accurate. With red wine, however, getting a truly representative juice sample

can be a challenge and can affect potential alcohol predictions. A juice sample taken soon after a tank is filled may not take into account un-popped berries, unripe berries (less sugar and more acids), and raisins (sometimes an overlooked source of large amounts of sugar, acid, and potassium). We suggest sampling after an initial 10°Brix drop, and analyzing the fermenting sample for glucose + fructose and alcohol simultaneously for a more accurate potential alcohol estimate. Proper sample preparation matters, too: in our lab, juices are centrifuged before analysis, and then mixed by inversion to avoid stratification, ensuring the most accurate results. Particulates have a minimal impact on refractometry, but can have a large impact on densitometry results.

Q : W H AT ' S T H E B E S T W AY T O P R E D I C T P OT E N T I A L A LCO H O L L E V E L S ? A:

Predicting potential alcohol levels in finished wines sounds simple, but there is more than one way to measure “sugar”, and formulas to convert this sugar into potential alcohol often miss the mark.

hydrometers or digital instruments), and other secondary measurements. The differences among the various measurement techniques are quite unpredictable depending on sample composition.

The “old school” method was to multiply °Brix by 0.6. One degree Brix is defined as 1 gram of sucrose in 100 grams of aqueous solution. However, grape juice does not naturally contain sucrose, but rather glucose, fructose, and a variety of organic acids and other dissolved solids. So when used for grape juice, °Brix is actually just an approximation of dissolved sugar, not an accurate representation of the fermentable sugars, and using ºBrix for estimating potential alcohol adds an additional layer of uncertainty to alcohol predictions. Differences between ºBrix and actual fermentable sugar content are even more pronounced in high ºBrix fruit and in fruit affected by fungal growth.

A more modern calculation that has proven to be more accurate uses glucose+fructose analysis, which provides a more accurate measurement of the levels of fermentable sugar compared to using ºBrix.

How ºBrix is measured also has an influence. Differences exist between ºBrix by refractometry, densitometry (using either

Note that in ripe fruit, glucose + fructose numbers often appear higher than the corresponding ºBrix results, because ºBrix is measured as a percentage by weight, meaning ºBrix values are greatly influenced by the density of the juice, while glucose + fructose is measured as weight by volume and is independent of juice density. An official conversion rate formula used in Europe is: Potential Alcohol (% vol) = glucose + fructose (g/L) / 16.83. In practice, rounding the 16.83 conversion factor to 17 is common.

29


SUGAR ANALYSIS In the wine industry, a term like "sugar" can mean different things. Clients often request testing for "RS", but this term can be very ambiguous.Â

30


R EDUCI NG S UGAR

°B R I X °Brix is a measurement of the apparent concentration of sugar. It is commonly used for grape juice and must and is expressed as a percentage by weight (% w/w). One degree Brix is defined as 1 gram of sucrose in 100 grams of aqueous solution. When the solution contains dissolved solids other than pure sucrose, as is the case for grape juice and must, the °Brix is only an approximation of dissolved sugar.

G LU COS E + F RU C TOS E In grape juice, Glucose + Fructose analysis measures the combined concentrations of the two main sugars present that can be consumed by yeast, also known as "fermentable sugars." Compared to °Brix, Glucose + Fructose can provide a better estimate of potential alcohol concentration after fermentation. In wine, “residual sugar” usually refers to the sum of Glucose + Fructose, an indication the amount of fermentable sugars remaining post fermentation, which is also an indication of ‘dryness’. Sucrose is not captured by this test. If it has been used in the winemaking process (such as for chaptalization of must, secondary fermentation of sparkling wine or added as a sweetener) measurement of Glucose + Fructose alone is usually not adequate – instead see Glucose + Fructose (Inverted)

Historically, “Residual Sugar” was measured by the Reducing Sugar method. This test derives its name from the ability of most sugars in juice or wine to ‘reduce’ other compounds. The most common reducing sugars are glucose and fructose. However, the method does not distinguish between fermentable and non-fermentable sugars, or other ‘reducing’ compounds for that matter, and these other compounds may contribute to reported results. Because of these limitations, the Reducing Sugar method is no longer the preferred choice to monitor completion of primary fermentation.

GLU COS E AND FR UCTOS E PANEL The Glucose and Fructose Panel provides the individual levels of glucose and fructose, in addition to their combined concentration. This test is often requested to investigate or remedy stuck or sluggish fermentations.

GLUCOS E + FR U CTOS E (I NV ER T ED) Inverted Glucose + Fructose provides the sum of the concentrations of glucose and fructose after “inversion” of the sample. Inversion is a process by which sucrose is broken apart into glucose and fructose, so that it can be measured and included in the reported results. Hence, this test is useful when “Residual Sugar” is required after chaptalization of must, secondary fermentation of sparklings or whenever sucrose has been used as a sweetener in wine, other alcohol beverages or spirits.

31


MONITORING GLUTATHIONE Glutathione is not an aroma compound itself, but is a powerful antioxidant that protects white wines and rosés from oxidation and loss of aroma or flavor. A low level of glutathione in grapes leads to lower levels in the juice, and early losses of aroma compounds. Glutathione, a natural grape antioxidant, can protect the aroma and flavor of white and rosé wines and prevents premature aging.

IBMP (2-Isobutyl-3methoxypyrazine) is the main compound responsible for the “green bell pepper” aroma in wine.

Glutathione levels fluctuate during production, as the compound can be absorbed by yeast and then released after fermentation. If final glutathione levels are low in young wines, the wines will experience faster loss of fresh varietal and fruity aromas, and poor aging potential. Monitoring glutathione levels can be beneficial through out the winemaking process to maximize white wine aroma and flavor, and prevent premature aging.

1 The glutathione content in grapes indicates their antioxidant potential, and can be influenced by a number of factors including soil nitrogen, vineyard practices, and grape maturity levels.

2 Eucalyptol- Eucalyptus traits are considered typical in some “cult” wines; on the other hand, in excess this character can be overwhelming. Even a slight “eucalyptus” note can interfere with delicate varietal aromas, and can have a detrimental influence on certain grape varieties..

Analyzing changes in glutathione levels during production helps to pinpoint where in the process glutathione is being lost – often from contact with air or exposure to copper residues.

3 A testing program can also identify winemaking processes that boost glutathione release after fermentation, and increase levels in wines.

A ROM A S 32


EUCALYPTOL

IBMP Isobutylmethoxypyrazine (IBMP) is the most important methoxypyrazine, a group of molecules responsible for very distinctive vegetal aromas in Sauvignon Blanc and a variety of red wines, mainly from the Cabernet family. In Sauvignon Blanc, these compounds add an often desired “grassy” character. In red wines however, the “green bell pepper” flavor is largely unpopular. Excessive IBMP levels lead to disappointing ratings and mixed success in the marketplace. The “green bell pepper” flavor in wine depends primarily on IBMP levels in harvested grapes. Once grapes have been picked, IBMP levels are not easily altered by standard winemaking processes.

Application: The intensity of “green bell pepper/grassy” characters in wines can be predicted by measuring IBMP in grapes right before harvest. Grape screening of IBMP helps identify “problem” vineyards or blocks. Since IBMP decreases during grape maturation, monitoring IBMP levels throughout ripening is a unique tool for assessing “aromatic maturity”. It allows targeting harvest dates based on desired aroma characteristics.

Since the discovery of eucalyptol (1,8-cineole) in red wines by ETS Laboratories in 2003, we’ve detected and measured eucalyptol in a large variety of wines exhibiting “eucalyptus-like” aromas. Flavors perceived during tasting were usually strongly related to concentrations of eucalyptol. Regardless of the grape variety, trace levels close to 1 ppb are associated with “fresh”, slightly “minty” notes. In the low ppb range, “minty” or “fresh bayleaf” aromas become stronger, and more easily identifiable as “eucalyptus” as concentrations increase. Wines with strong “eucalyptus” odors may contain more than 20ppb of eucalyptol. Eucalyptol’s sensory impact in wine is considered more or less desirable depending on the grape variety. Wines from southern Rhône and Mediterranean varieties seem better able to accommodate the characteristic well. Moderate levels can be appreciated in wines from Bordeaux varieties. Whereas with Pinot Noir, even trace levels can detract from varietal expression.

Application: As eucalyptol may be mostly contributed by eucalyptusderived MOG (leaves, bark debris…) hiding in grapes, analyzing grape samples before harvest is mostly pointless, and routine testing is not offered by ETS Laboratories. On the other hand, wine analysis assists winemaking teams in objectively documenting their sensory impressions and managing a strong flavor component. Unlike IBMP and smoke-derived compounds (see p. 38), eucalyptol is relatively slow to extract during red winemaking, making shorter macerations a valid strategy to minimize impact. Winemakers who wish to minimize or maintain consistent levels of “eucalyptus” character will also benefit by determining eucalyptol concentrations in distinct wine lots prior to blending.

Monitoring IBMP from the early stages of the ripening process can greatly improve fruit quality from underperforming vineyards. Levels in grapes are well known to be linked to vine vigor, canopy and water availability, with severe heat occasionally causing IBMP’s natural degradation to stop. Once the kinetics of IBMP accumulation and degradation in specific sites are understood, viticultural practices can be modified accordingly. Once grapes have been picked, IBMP levels are not easily altered by standard winemaking processes. The IBMP potential of grapes can be grossly underestimated from juice samples, making whole berries the preferred sample in most cases. Analyzing juice samples may be relevant in white winemaking, however. 33


ETS APPLIED RESEARCH:

THE IMPAC T OF SMOKE 17 GRAPE SAMPLE S FROM MULTIPLE FIRE EVE NTS

GRAPE TESTS FOR SMOKE IMPACT

More recently, small scale fermentations, also called “bucket ferments" or “microferments, followed by sensory evaluation and analysis of fermented samples, have been proposed for risk assessment (4). This work shows the relationship between “free” guaiacols measurable in red grapes, and guaiacols found in microferments or in commercial scale wines, from three recent vintages in California. 2015 Vintage: Guaiacols Pre and Post Acid Treatment of Red Grapes, and in Microferments Wildfires impacted several California viticulture areas during the 2015 vintage. We could obtain fruit samples from multiple red grape varieties exposed to smoke from several separate fire events. After destemming, berries were split in three aliquots. One aliquot was homogenized and analyzed directly for free guaiacols using headspace/SPME/GC/MS. A second homogenate was submitted to an acid and heat treatment (pH = 1, 95°C, 1h ), which had been proposed to hydrolyze glycosylated guaiacols (5), prior to analysis. Berries from the third aliquot were crushed, inoculated and fermented in contact with their skins and seeds in small containers. Musts obtained were mixed daily during fermentation. After primary fermentations were completed, the red wine “microferments” obtained were drained and analyzed for guaiacols using Headspace/SPME /GC/MS (Fig. 1).

34

ANALYSIS AFTER ACID +HEAT TREATMENT

MICROFERMENTS AND WINE ANALYSIS

FIGURE 1: OUTLINE OF AN EXPERIME NT WI TH GRAPE S EXPO SED TO SMOKE FROM SEVERA L WILD FIRE S IN VARI OUS CA LIF ORNIA VI TIC ULT URA L AREAS DURING THE 201 5 VINTAGE.

Free guaiacol levels in grape homogenates were correlated with levels found in red microferments (R2 = 0.743). Similar results were observed with 4-methylguaiacol (data not shown). Surprisingly, guaiacol levels measured after acid and heat treatment of the homogenates were only slightly higher than free levels measured without treatment. They did not show any improved correlation with levels in microferments (R2 = 0.743) (Fig. 2) . 45.0

Y=2.3514X

40.0 WINE FREE GUAIACOL

During fermentation, however, part of these glycosylated compounds appear to be hydrolyzed, giving back odor-active “free” volatiles.

ANALYSIS OF INTACT BERRIES

2

R =0.7433

45.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0 18.0

20.0

16.0 18.0

20.0

BERRY FREE GUAIACOL 45.0

Y=2.0489X

40.0 WINE FREE GUAIACOL

The analysis of “free” guaiacol (and 4-methylguaiacol) in smoke-exposed grapes has been proposed more than ten years ago as a way to predict the risk of smoke characters in wines (1, 2). A main concern, however, is that a significant part of smoke volatiles in grapes, including guaiacols, become glycosylated (3) and evade detection by GC/MS.

2

R =0.7272

45.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

BERRY HYDROLYZED 1 hGUAIACOL

FIGURE 2: LEF T: FREE GUAIAC OL IN HOMOGE NATE S FROM 201 5 RED GRAPE S EXPO SED TO WILD FIRE SMOKE, C OMPARED TO GUAIAC OL MEASURED IN MIC ROFERME NTS FROM THO SE GRAPE S. RIGHT: GUAIAC OL IN THE SAME HOMOGE NATE S AF TER HEAT AND AC ID TREATME NT (PH = 1, 95°C, 1H ), C OMPARED TO GUAIAC OL IN THE SAME MIC ROFERME NTS.


45.0

X5

40.0 WINE FREE GUAIACOL

Guaiacol levels in microferments appeared usually approximately 3 to 5 times higher than free guaiacol levels measured in grape homogenates, confirming the cleavage of precursors during fermentation. An interesting exception was with grapes exposed to smoke only 3 days before harvest. In that case free guaiacol barely increased post-fermentation (Fig. 3). It is possible that exposure to smoke late in the season allowed less time for modifications of smoke volatiles, e.g. by glycosylation, to happen, or/and that such modifications occur less in senescent berries.

X3

45.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0 18.0

20.0

BERRY FREE GUAIACOL FIGURE 3: FREE GUAIAC OL IN RED GRAPE S EXPO SED 201 5 WILD FIRE S VS. GUAIAC OL IN MIC ROFERME NTS, INDICATING A THREE TO FIVE F OLD INC REASE IN MO ST CASE S. THE C IRC LED OUTLIER I S FROM GRAPE S EXPO SED TO WILD FIRE SMOKE ONLY THREE DAYS BEF ORE HARVE ST.

Smoke impact is caused by a wide range of volatile

phenols found in wildfire smoke. These compounds are absorbed by vines and accumulate in berries. They eventually end up in wine where they can cause unwanted flavors. These off-flavors, described as “smoky”, “bacon”, “campfire” and “ashtray”, are usually long lasting and linger on the palate even after the wine is swallowed or spit out. Smoke impact in wine was identified as a serious problem after the 2003 wildfires in Australia and British Columbia. The California wine industry was also affected following the wildfires of summer 2008, and smoke taint has been a concern for many growers and wineries ever since.

Application: Two of the main volatile phenols in smoke, guaiacol and 4-methylguaiacol, are useful markers of smoke impact in wines. Their concentration is usually correlated with the degree of perceived smoke impact, particularly in wines not exposed to toasted oak. During the 2008 California wildfires, ETS developed an analytical tool to screen grapes for the risk of smoke impact. The analysis measures trace levels of free guaiacol and 4-methylguaiacol in whole berries. Ten vintages later, we are very confident that knowing the levels of these indicator compounds in berries enables winemakers to assess the risk of smoke impact, and chose an appropriate course of action to mitigate the effects in their wines.

1. Berries

2. Juice

3. Wine

Exposure of vines to smoke can be quite variable in a given vineyard, and getting a representative sample can be challenging. Berries should be submitted undamaged as much as possible and kept cold with icepacks during shipment. It is preferable to collect samples close to the harvest date.

It is possible to measure smoke taint markers in juice samples, but since smoke compounds are mostly located in skins, whole berry testing is the preferred method for pre-harvest screening.

Analyzing wine samples is a useful tool for confirming perceived sensory faults. It is preferable to sample and analyze wine that has not come in contact with toasted oak. Testing is usually required to help make decisions regarding taint removal treatments. 35


FREE GUAIACOL IN WINE

(u g / L )

X3

40

35

M U LT I P L I E R

30

Y=3.5968X

25

2

R =0.9013

20 15 10 5 0

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

(u g/ kg )

Discussion X1

30 20 10 0

5

10 15 20 25 FREE GUAIACOL IN BERRIES

30 35 ( u g /k g)

40

FIGURE 4: FREE GUAIAC OL IN GRAPE S EXPO SED TO SMOKE FROM THE OC TOBER 2017 WINE C OUNTRY FIRE S (79 SAMPLE S) C OMPARED TO GUAIAC OL MEASURED IN C ORRE SPONDING PRODUC TI ON RED WINE S PO ST PRIMARY FERME NTATI ON, INDICATING AT THE MAXIMUM A THREE F OLD INC REASE BE T WEE N GRAPE AND WINE LEVELS.

2018 Vintage: Free Guaiacols in Cabernet Sauvignon grapes and in Microferments The Mendocino Complex Fire occurring from late June to early November 2018 prompted the Lake County Winegrape Commission to initiate a study comparing grape tests with microferments (6). Cabernet Sauvignon samples representing a wide range from low to high exposure to smoke were collected in September, and split for immediate analysis and microfermentations. Free guaiacol in grapes were well correlated with guaiacol in microferment, with microferment levels usually three to five times higher than grape levels (Fig. 5), rather in line with our 2015 observations.

36

40

FIGURE 5: FREE GUAIAC OL IN CABERNE T SAUVIGNON GRAPE S EXPO SED TO SMOKE FROM THE 201 8 ME NDOC INO C OMPLEX FIRE, C OMPARED TO GUAIAC OL IN MIC ROFERME NTS FROM THO SE GRAPE S

50

0

2018 L A K E CO U N T Y C S G UA I ACO L I N M I C RO F E R M E N T S VS. B E R R I E S

BERRIES GUAIACOL

2017 N A PA F R E E G UA I ACO L W I N E VS. G R A P E S

60

( u g /L )

The Wine Country Fires occurred at the tail end of the 2017 vintage in the Napa and Sonoma viticulture areas, impacting grapes, mostly from red varieties, that hadn’t been harvested yet. A comparison of free guaiacol in berry samples submitted at ETS Laboratories, and in corresponding wines (all produced on a commercial scale) analyzed post primary fermentation (Fig. 4), shows that wine levels were only approximately three times higher than grape levels. This again might be explained by less time for modifications of smoke volatiles and/or lessened modification activity in senescent grapes.

MICROFERMENTS GUAIACOL

2017 Vintage: Free Guaiacols in Red Grapes and in Red Wines

Based on results obtained during three recent vintages in California, analyzing free guaiacol (and 4-methylguaiacol) in grapes is confirmed to be a valid test for assessing smoke-exposed grapes. While microferments can undergo sensory evaluation and probably offer a more direct and accurate prediction of guaiacol levels in production wines, they have the inconvenience of taking several days to prepare. Pros and cons of both approaches can be summarized as follows:

GRAPES

M icro fe r m e n t s

Pre pa ra ti on T i me ( be fore s e n di n g s a mple s to th e la b ora tor y)

I mme di a te

> 1 we e k

S e n s or y Eva lua ti on

N ot ve r y us e f ul

Us e f ul , b u t d ifficu l t ( n e e d for m u l t ip l e t rain e d t as t e rs )

An a lys i s Turn a roun d T i me

1- 2 da ys

1 - 2 d ays

Pre di c ti on of G ua i a col Leve ls i n Prod uc ti on W i n e s

I n di re c t ( va ri a ble " multi p li e rs " be twe e n g ra p e a n d wi n e re s ults

Re d s : m o re d ire ct , b u t d e l aye d W h i t e s : u n ce r t ain ( fe rm e n t w it h s k in s fo r " worst cas e s ce n ar io s " ? )

REFERE NC E S 1. WHI TING J. AND KR STIC M. 2007. IMPAC T OF BUSHFIRE SMOKE ON GRAPE AND WINE, ST UDY REPOR T, DEPT OF PRIMARY INDUSTRIE S, VIC TORIA, AUSTRA LIA. 2. HERVE, E.; PRIC E, S. AND BURNS, G. FREE GUAIAC OL AND 4-ME THYLGUAIAC OL AS MARKER S OF SMOKE TAINT IN GRAPE S AND WINE S: OB SERVATI ONS FROM THE 2008 VINTAGE IN CA LIF ORNIA. 2011.PROC EEDINGS OF THE 9E SYMPO SIUM INTERNATI ONA L D’Œ NOLOGIE, BORDEAUX, FRANC E, 3. GLYC O SYLATI ON OF SMOKE-DERIVED VOLATILE PHE NOLS IN GRAPE S AS A C ONSEQUE NC E OF GRAPEVINE EXPO SURE TO BUSHFIRE SMOKE HAYASAKA Y.; BA LDOC K G.; PARKER M., PARDON K., BLAC K C., HERDERIC H M., AND JEFFERY D2010. JOURNA L OF AGRIC ULT URA L AND F OOD C HEMI STRY 58 (20), 10989-10998. 4. THE AUSTRA LIAN WINE RE SEARC H INSTI T UTE. FAC T SHEE T “SMA LL LOT FERME NTATI ON ME THOD” FAC T SHEE T . UPDATED MAY 201 9. 5. SINGH P., C HONG H., PI TT M., C LEARY M, DOKOOZLIAN N., DOWNE YSINGH D. AND C HONG H. GUAIAC OL AND 4-ME THYLGUAIAC OL ACC UMULATE IN WINE S MA DE FROM SMOKE-AFFEC TED FR UI T BECAUSE OF HYDROLYSI S OF THEIR C ONJUGATE S. AUSTRA LIAN JOURNA L OF GRAPE AND WINE RE SEARC H. 17. S1 3 - S21.2011. 6. MCGOUR T Y G, JONE S M, KEIFFER R. 201 9. MIC ROVINIFICATI ONS AS A ME THOD F OR PREDIC TING SMOKE TAINT IN WINE. ASEV NATI ONA L C ONFERE NC E.


37


Fermentation #1: 2016 Merlot – Carmel Valley, California Grape Variety: Merlot

RED WINE MACERATION TIME AND SMOKE IMPACT Smoke volatile compounds from wildfires can be absorbed by vines and grapes and cause off-flavors in wines. Once absorbed by grapes, smoke derived compounds appear to be mostly localized in skins. Consequently, white grapes pressed with minimal skin contact give wines with less smoke compounds than when more skin contact takes place. Fermentation of red grapes, in presence of skins, gives even higher levels (1, 2). These observations have been often interpreted as a reason to shorten skin contact in red winemaking, despite the lack of studies actually tracking smoke markers during red wine maceration and fermentation. Alas, winemakers trying this approach often report it fails to mitigate smoke impact in red wines, at the cost of changing often drastically their phenolics composition. To investigate the extraction of smoke derived compounds in red winemaking, juice and fermenting juice samples from smoke-exposed red grapes were taken from fermenters at different stages. Samples were not analyzed immediately. Instead, fermentations were initiated and/or completed in glass bottles, without any further contact with skins. Doing so allowed the partial hydrolysis of glycosylated smoke phenols, known to occur during fermentation (3), to take place. The resulting samples were wines displaying colors ranging from light rose to dark red. They were analyzed for total anthocyanins, tannins (defined as proanthocyanidin polymers of at least four flavan-3-ol units) and catechin using reversed-phase HPLC (4). The smoke impact was assessed by measuring free guaiacol and 4-methylguaiacol using Headspace/SPME/GC/MS/ MS.

38

Origin: Carmel Valley, California Fire Event: Soberanes fire, July 22, 2016 – October 12, 2016 Harvest: manual on October 8, 2016, 100% destemming Fermentation: stainless steel tank, four days of cold soak, moderate fermentation temperature Draining and Pressing: at dryness after 13 days Extraction of all measured polyphenols was slow during cold soak and then followed predictable patterns (figure 1). After the start of fermentation, anthocyanins were extracted rapidly during two days, then their concentration levelled off. Tannin extraction mimicked anthocyanins extraction at first, suggesting a fast extraction of skin tannins, but levels kept increasing significantly until draining and pressing. This latter phase can be explained by a continuous extraction of seed tannins, as indicated by the steady release of catechin, an indicator of seed phenolics. By contrast, guaiacol and 4-methylguaiacol, the two markers of smoke impact, reached virtually their maximum in wines obtained from juice sampled after only three days, still at a rosé stage (4-methylguaiacol was ten times less abundant than guaiacol in all samples obtained from this ferment – data not shown). M E R LOT 2016 S O B E R A N E S F I R E

700 600 500 400

CAT X10 300 TANNIN 200 TOTA L A 100 G X 10 0

0

2

4

6

8

10

12

14

FIGURE 1: EXTRAC TI ON OF POLYPHE NOLS (MG/L TANNINS, TOTA L ANTHOCYANINS, CATEC HIN X10) AND THE SMOKE MARKER GUAIAC OL (UG/L X10) DURING THE FERME NTATI ON AND MAC ERATI ON OF A 201 6 CARMEL VA LLE Y MERLOT.


Fermentation #2: 2017 Cabernet Sauvignon – Sonoma Valley, California Grape Variety: Cabernet Sauvignon

Fermentation #3: 2017 Petit Verdot – Napa Valley, California Grape Variety: Petit Verdot Origin: Napa Valley, California

Origin: Sonoma Valley, California Fire Event: Wine Country Fires October 8, 2017 – October 31, 2017 Harvest: manual on October 23, 2017, 100% destemming

Fire Event: Wine Country Fires October 8, 2017 – October 31, 2017 Harvest: manual on October 24, 2017, 100% destemming Fermentation: stainless steel tank, no cold soak, warm fermentation temperature

Fermentation: stainless steel tank, no cold soak, moderate fermentation temperature

Draining and Pressing: at 11 degrees Brix after 4 days

Draining and Pressing: at dryness after 13 days Anthocyanins steadily increased during the first nine days then levelled off (figure 2). Tannin extraction followed anthocyanins extraction for seven days then increased faster, which again can be explained by the extraction of seed phenolics, as indicated by the release of catechin. Guaiacol and 4-methylguaiacol reached their maximum in wines obtained with juice sampled after only two days extremely early in the fermentation process (4-methylguaiacol was five times less abundant than guaiacol in all samples obtained from this ferment – data not shown).

In this third example, the winemaker tried a warm and fast fermentation with draining and pressing of the partially fermented must after four days. Anthocyanins increased quickly during the first three days then levelled off (figure #3) while tannin levels kept increasing. Unfortunately, guaiacol and 4-methylguaiacol had reached their maximum one day before draining and pressing (4-methylguaiacol was five times less abundant than guaiacol in all samples obtained from this ferment – data not shown).

C S 2017 S O N O M A

100% P E T I T V E R D OT N A PA VA L L E Y 2017 (D R A I N E D @ 11 B R I X)

1400

1 600

1200

1400

1000

1200 1000

800 600 400

800

TANNIN

600

TOTA L A

200 0

CAT X10

G X 10 0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

FIGURE 2: EXTRAC TI ON OF POLYPHE NOLS (TANNINS, TOTA L ANTHOCYANINS, CATEC HIN X10, IN MG/L) AND THE SMOKE MARKER GUAIAC OL (UG/L X10) DURING THE FERME NTATI ON AND MAC ERATI ON OF A 2017 SONOMA VA LLE Y CABERNE T SAUVIGNON.

CAT X10 TANNIN TOTA L A

400

G X 10 200 0

0

1

2

3

4

5

6

7

8

9

10

11

12

FIGURE 3: EXTRAC TI ON OF POLYPHE NOLS (MG/L TANNINS, TOTA L ANTHOCYANINS, CATEC HIN X10) AND THE SMOKE MARKER GUAIAC OL (UG/L X10) DURING THE FERME NTATI ON AND MAC ERATI ON OF A 2017 NAPA VA LLE Y PE TI T VERDOT.

Discussion

tannin levels are achieved.

Based on results obtained for free guaiacol and 4-methylguaiacol, two major markers of smoke impact in wine, the full extraction of smoke-derived volatile compounds in red ferments appears to be achieved extremely quickly, often before full anthocyanins extraction, and before desirable

The consequence is that shortened macerations seem futile in mitigating the impact of wildfire smoke in red wines. The more radical approach of making rosé wine may be also often ineffective, unless a “blanc de noir” method with no skin contact at all is applied.

REFERE NC E S 1. KELLY D, ZERIHUN A, HAYASAKA Y, GIBBERD M. 2014. WINEMAKING PRAC TIC E AFFEC TS THE EXTRAC TI ON OF SMOKE-BORNE PHE NOLS FROM GRAPE S INTO WINE S AUSTRA LIAN JOURNA L OF GRAPE AND WINE RE SEARC H. 20 (3), 386-393. 2. KE NNI SON KR, GIBBERD MR, POLLNI T Z AP, WILKINSON KL. 2008. SMOKE-DERIVED TAINT IN WINE: THE RELEASE OF SMOKE-DERIVED VOLATILE PHE NOLS DURING FERME NTATI ON OF MERLOT JUIC E F OLLOWING GRAPEVINE EXPO SURE TO SMOKE. JOURNA L OF AGRIC ULT URA L AND F OOD C HEMI STRY. 56 (1 6), 7379-7383. 3. RI STIC R, O SIDAC Z P, PINC HBEC K A, HAYAKASA Y, FUDGE A L, WILKINSON KL. 2011. THE EFFEC T OF WINEMAKING TEC HNIQUE S ON THE INTE NSI T Y OF SMOKE TAINT IN WINE. AUSTRA LIAN JOURNA L OF GRAPE AND WINE RE SEARC H 17, S29-S40. 4. WATERHOUSE A, PRIC E S AND MCC ORD J. 1 999. REVER SED-PHASE HIGH PRE SSURE LIQUID C HROMATOGRAPHY ME THODS F OR ANA LYSI S OF WINE POLYPHE NOLS. ME THODS IN E NZYMOLOGY 299, 11 3-121.

39


40


NE W TO ETS :

S MOK E MARKERS PA N E L S Sm oke Vo l ati le Markers - Ex p and ed Panel New for 2019, ETS is now offering an extended panel of smoke volatiles in their “free” forms (meaning not glycosylated = bound to sugars), using Headspace/SPME/GC/ MS/MS (QQQ). This panel includes guaiacol, 4-methylguaiacol, o-cresol, m-cresol, p-cresol, phenol, syringol and methylsyringol. This expanded panel complements our standard Smoke Volatile Guaiacols Panel, comprised of guaiacol and 4-methylguaiacol.

Sm oke Glycosyl ated Markers Panel ETS has developed a unique and very inclusive panel of glycosylated smoke markers using a state-of-the-art combination of solid phase extraction, liquid chromatography, and triple quadrupole mass spectrometry (SPE/HPLC/MS/MS QQQ). This comprehensive panel will provide unmatched insight into the “hidden” forms of smoke volatiles absorbed by grapes and then becoming bound to a variety of sugars, making them odorless. Glycosylated (or bound) smoke compounds are only partially hydrolyzed (or released) during fermentation. They have been described to contribute the disagreeable and lingering “ashy” aftertaste often experienced with smoke-impacted wines. They are also suspected to slowly hydrolyze in wine, releasing little by little odor-active free compounds, which may with time increase the sensory perception of smoke characters. Our goal is to offer this panel during the 2019 harvest.

41


MO N ITO R I N G RISK FACTO R S FO R QUE R C E T I N PRECIP I TAT ES Q u e rce tin p re c ipita tes in bottled w ines are par t i c ular ly nas t y as a c lear, a pp a re ntl y s tab l e w ine a t bo ttling ca n fo r m a greeni s h goopy prec i pi t at e a f t e r b ottl ing aging. Precipita tes ca n fo r m i n les s t h an a year i n bot t le b u t a re more common two to fo ur ye ars aft er bot t li ng. Oft en t h e p re c i p itation d oes no t occur until a fter t h e wi ne i s on t h e m ar ket . T h ere ha ve b e e n mu l tiple reca lls o f problem wi nes from t h e m ar ket .

42


T

OH

he underlying issues were discussed in a previous issue (Winemaker's Quarterly Vol. 3 Issue 2) but in summary the problem is caused by flavonols glycosides (quercetin, myriceitin, kaemferol and isorhamnitin OH glucosides and glucuronides) hydrolyzing to far less soluble aglycones. Tannin and anthocyanins help keep these compounds in suspension in wine but, as wines age, anthocyanin content decreases and flavonols glycosides continue to hydrolyze and add to the pool of less soluble aglycones. Quercetin is the most common flavonol in grapes and wine and precipitatesHO seen in the lab tend to be mostly O quercetin aglycones. The problem is very common in Sangiovese due to its genetic propensity to synthesize quercetin glycosides and its low anthocyanin and tannin concentrations. However, quercetin precipitates are becoming increasingly common in Merlot and other lower tannin red wines. ETS Laboratories has several analytical tools that may help identify risk factors for quercetin precipitation. Flavonol O precipitates can be identified as part of a Sediment Evaluation. Analysis of grapes with the ETS Grape Phenolic OH PanelOcan quantify quercetin glycoside levels in grapes coming into the winery and is useful to evaluate vineyard treatments intended to lower flavonol synthesis. The assay includes OH quercetin glycosides as well as tannin and total anthocyanins.

O

The ETS Red Wine Phenolic Profile is a detailed look at a wide range of phenolic components in wine and is useful in identifying potential risk as well as evaluating winery practices intended to reduce quercetin in wine. Treatments that are typically used to stabilize wines usually include blending with high tannin wines and PVPP fining prior to bottling. The Red wine Phenolic Profile includes tannin, total anthocyanins, quercetin glycosides and quercetin aglycones, all potentially important in understanding the potential stability of a wine.

HO

ETS staff are always available to help with interpretation of results.

HO OH

43


HARVEST TOOLKIT Get the most out of your harvest.

G R AP E M ATU RI TY M ON I TORI N G PAN ELS The ETS Grape Maturity Monitoring Panels provide a set of analyses that growers and winemakers request to monitor fruit maturity, including Brix, Glucose + Fructose, TA, pH and malic acid, as well as berry size parameters and average sugar per berry. Monitoring sugar per berry allows growers to determine the duration and rate of active sugar loading, during which vines synthesize and actively transport sugar into berries. The time at which the sugar loading phase ends, and which Brix was achieved at that time, are both important indicators of wine growing conditions (excessive or insufficient vigor, water availability and resistance to heat stress‌). For more details see pg. 4

44


E T S S CORP I ON S

TM

Wild yeast and bacteria from the vineyard may be introduced into the winery on the harvested fruit, causing spontaneous fermentation and spoilage. ScorpionsTM DNA analysis offers winemakers an early detection tool to identify these spoilage organisms. Despite the best practice of modern winemaking methods, microbial contamination often occurs during wine production. Spoilage microbes are capable of survival and growth in the wine, potentially producing off-flavors, off aromas, and turbidity. Microbiological contamination is often undetected until related problems in the wine become noticeable by sensory evaluation.

Scorpions™ assays, based on specific genetic targets, detect the full range of wine and juice spoilage organisms. This genetic analysis method detects microbial populations directly in wine or juice. Results are routinely reported within two business days, giving winemakers the ability to address problems before wine defects occur. Targeted genetic probes give the winemaker the ability to monitor only those specific spoilage organisms that have the potential to adversely impact wine quality, and to accurately measure populations down to extremely low levels. For more details see pg. 20

45


H A R V E S T TOOLKIT

GRA PE P H E N O LI CS Follow changes in grape phenolics during ripening, using the catechin and tannin "ripeness" index to monitor seed ripening, and quercetin analysis to monitor canopy effects on grape phenolics. For more details see pg. 24

GRAP E WATER CON TEN T Changes in grape water content influence finished wine composition and can be as important as standard sugar and acid measurements when making picking decisions. Grape water content is also very useful for understanding changes in TA, pH, ยบBrix, and other harvest indicators.

MO N I TO R I N G I B M P The compound responsible for a "green bell pepper" aroma in wine, IBMP, decreases quickly during maturation. But once grapes are picked, it is hard to control. Monitoring changes in grape IBMP directly influences final levels in wine, and is crucial in making picking decisions. For more details see pg. 33

46


H A R V E S T TOOLKIT

GLU TAT H I O N E Glutathione, a natural grape antioxidant, can protect the aroma and flavor of white and rosé wines and prevent premature aging. Glutathione levels fluctuate during production, as the compound can be absorbed by yeast and then released after fermentation.

For more details see pg. 32

S M OKE I M PACT The compounds in smoke are absorbed by vines and can cause unwanted flavors in wine. Analyzing for these compounds allows winemakers to screen grapes for the risk of smoke impact and work to mitigate its effects. For more details see pg. 34

HA R VE S T J U I CE PA N E L Our most popular harvest panel offers a full range of grape and must analyses, combining more than 10 tests including fermentable sugar (to help estimate alcohol content) and YAN (yeast-assimilable nitrogen ­­— to help predict sluggish or stuck fermentation and potential sulfide formation.) For more details see pg. 7

47


H A R V E S T TOOLKIT

B OTRY TI S PAN EL This comprehensive test panel checks grapes for Botrytis (using ScorpionsTM ) and laccase detecting both the spoilage organism and its byproducts that can harm your wine.

L ACCA S E AC T I VI T Y Laccase is a polyphenol oxidase associated with rot caused by Botrytis. Elevated levels of laccase can result in oxidation of phenolic compounds that may cause color degradation or premature browning in red wines. In addition, laccase mediated oxidation can also affect the aroma profile of the wine.

Y EAS T V I AB I LI TY Our automated method reports yeast viability and total cell count within hours, and the real-time microscopic flow image analysis examines 1,000 times the volume used in standard microscopic methods, vastly increasing the accuracy of your results.

48


H A R V E S T TOOLKIT

R A P I D PH E N O L I C S By the end of maceration or fermentation, the tannin content of a wine is already fixed. Monitoring phenolics during this critical period allows winemakers to better control tannins by increasing or decreasing phenolic extraction. For more details see pg. 25

DN A FI N GERP RI N TI N G ETS Laboratories offers DNA fingerprinting to distinguish between closely related strains of Saccharomyces cerevisiae. ETS MLVA technology allows winemakers to monitor yeast population in native fermentations and check the efficiency of inoculations with commercial strains.

E U C A LY P TO L Eucalyptus character is a controversial sensory expression in red wines from California and countries with Mediterranean climates. Even a slight “eucalyptus� note can interfere with delicate varietal aromas, and can have a detrimental influence on certain grape varieties. For more details see pg. 33

49


HARVEST

RESOURCES

Since we first opened our doors in St. Helena in 1978, we’ve grown alongside the wine industry, partnering with our clients as they’ve gone on to create many of the world’s finest wines. We’re continuing to invest not only in Napa Valley, but also in other rapidly growing wine regions by expanding our local and online services to support winemakers with advanced tools and technical assistance. Here’s what’s new for harvest 2019:

Our New Space

Over the last year we've been expanding our St. Helena lab and we are excited to have our new space ready for Harvest 2019.

Our Cross Client Search Feature is now live! If you are a team member with more than one client, you can now search through all of their results using one tool. Instead of having to drill down into each client separately and search through only that client’s results, you can now access all of your data with a quick search. Our new search tool works exactly the same as the current client test result feature, but now you are able to select and deselect which client test results to include in the results displayed.

We've doubled our lab space allowing us to add more instruments, new analyses and meeting space. Our team is very excited about the extra elbow room and we hope that you are able to stop in this Harvest to see our updated laboratory.

INTRODUCING THE NMR WIN E AUTHENTICITY

ANALYSIS PRINCIPLE

We are excited regarding our recently installed Wine Screener which provides comprehensive authenticity and quality control screening using Proton Nuclear Magnetic Resonance Profiling (1 H-NMR).

1 H-NMR analysis generates a spectrum representing the unique fingerprint of the tested wine. The profile provides a large range of information both targeted, for quantification of interest compounds and untargeted, that might indicate contamination or reveal anomalies in the sample.

The technique and databases have been developed in partnership with Bruker Biospin and a network of partners supplying authentic samples from Europe and some regions in South America. ETS will be working with Bruker to build solid and reliable databases for wines from the Americas. 50

ADVANTAGES OF THE METHOD

This screening method covers a large scope of potential issues including detection of counterfeiting, mislabeling, or quality deviations

Current models are in place for discrimination of grape variety for pure varietals, geographical origins, and other issues in Europe and Chile.


L O C A L J U I C E A N A LY S E S AT S AT E L L I T E L A B O R AT O R I E S A L L S AT E L L I T E S

Samples can be dropped off locally for any analysis ETS offers. These common harvest analyses will be run onsite for quicker turnaround.

A N A LY S I S

TECHNIQUE

TURNAROUND

JUICE PANEL

MINERVA

SAME DAY

BUFFER CAPACITY

MANUAL

SAME DAY

ETHANOL

MINERVA

SAME DAY

FREE SO2

FLOW INJECTION

SAME DAY

TOTAL SO2

FLOW INJECTION

SAME DAY

TURBIDITY

TURBIDIMETRY

SAME DAY

VOLATILE ACIDITY

SEQUENTIAL ANALYZER

SAME DAY

GRAPE MATURITY MONITORING PANEL

VARIOUS

SAME DAY

RAPID PHENOLIC PANEL

HPLC

1 DAY

GRAPE PHENOLIC PANEL

HPLC

1 DAY

SCORPIONS BACTERIA JUICE PANEL

SCORPIONS™

2 DAYS

SCORPIONS YEAST JUICE PANEL

SCORPIONS™

2 DAYS

SCORPIONS COMBINED JUICE PANEL

SCORPIONS™

2 DAYS

BRIX GLUCOSE + FRUCTOSE PH TA (TITRATABLE ACIDITY) TARTARIC ACID L-MALIC ACID POTASSIUM NOPA AMMONIA

PA S O R O B L E S , WA L L A WA L L A & NEWBERG

PA S O R O B L E S & NEWBERG

VISIT OUR WEBSITE FOR CURRENT PRICES, AND THE FULL LIST OF O N S I T E W I N E A N A LY S E S .


H A NDLE W I T H

CARE

It’s important to collect and handle harvest samples carefully to ensure accurate and representative results. We’ve collected our recommendations to help you get the most out of your harvest analyses.

JUICE SAMPLING Most harvest samples received at ETS come to the laboratory as juice. Berries pressed for a juice sample should be selected from at least 20-40 different clusters, and can easily be pressed by hand in their collection bag – pour the juice into a standard ETS 60mL sample tube and label with your ETS client labels. Samples should be kept cool to prevent fermentation.

GRAPE SAMPLING FOR PHENOLICS For grape phenolic testing, a representative sample is critical to obtain accurate results, especially in varietals with tight clusters. Samples for the Grape Phenolic Panel should include berries from at least 20-40 different clusters. Clusters can be collected either from harvest containers or directly from the vineyard. To get a representative sample, all the berries must be stripped from the clusters and mixed before bagging a 300-400 berry sample for analysis (about 500g or 16 oz). Samples should contain only intact and undamaged fruit to ensure accurate results.

BERRY SAMPLING Take 200-400 berries per block. Pick berries from random clusters on both sides of the row. Take berries from the top and bottom of both the front and back of each cluster. *Samples submitted for berry analysis should contain only intact and undamaged fruit.

We encourage clients to submit berry samples rather than whole clusters. If samples are submitted as clusters, ETS will prepare a berry sample for an additional fee.

LABELING Each bag of berries should be clearly labeled with the client name, sample ID, and analyses to be performed. ETS provides free sample labels that are pre-printed and barcoded with your client ID – visit our website, or give us a call.

52


GEAR UP FOR HARVEST

Don’t get caught empty handed – order complimentary tubes, pre-printed labels, and shipping pods to take the headache out of collecting and shipping samples. 1. Login to your ETS account and select the winery you're ordering supplies for.

Outside the range of our sample pickup service? ETS offers free shipping kits, which include an insulated envelope and an ice pack, to help you easily send in samples no matter where you're located.

S H I PP IN G JU I CE S AM P LES FO R R EGU LAR AN ALYS ES

S HI P P I N G JU I CE SAMPLES FOR S CORP I ON ANALYSIS

To prevent problems from fermentation, juice samples should be frozen or boiled for shipment, and clearly marked as either “FROZEN” or “BOILED” depending on the treatment used.

To ensure accurate results, it's important to avoid damaging DNA or killing yeast and bacteria:

Boiling 2. Use the "Get Supplies" button on the dashboard to place a supply order.

Boil samples with a loosely fitting cap to prevent evaporation and concentration. Do not over-boil.

° Keep samples cool with ice packs. Ship by overnight delivery using a °parcel carrier like FedEx, UPS, or GSO. Scorpions samples should not be frozen or boiled.

Freezing Freeze the sample in a plastic ETS sample tube. Do not over-fill the tube – leave a small space for the sample to expand when frozen. Never freeze samples in glass containers to prevent breakage and injury. 53


L O C AT I O N S & SERVICES In addition to our St. Helena headquarters, we operate satellite laboratories across the West coast to bring advanced tools to winemakers' doorsteps and provide local support to other growing wine regions. These quick guides provide a reference for the 2017 harvest, including weekend hours and convenient services to make it easier than ever to send in harvest samples, including dropbox locations and complimentary courier service. As always, if you have any questions, just call your local lab and we'll be happy to help.


S T. H E L E N A C A L I F O R N I A

A F T E R H O U R S D RO P B OX S T. H E L E N A

PHONE

(707) 963-4806

Located at our laboratory, next to the main entrance.

COURIER SERVICE Our complimentary courier service is available in Napa, Sonoma, and Mendocino counties every day ETS is open.

Samples left overnight will be processed when we open the following business day.

SHIPPING ADDRESS

REQUEST A PICKUP

899 Adams Street, Suite A St. Helena, CA 94574

You can request a courier pickup by logging in to your ETS account, or by calling our lab.

SEE INSTRUCTIONS FOR SHIPPING JUICE S A M P L E S – P. 4 1 DEADLINE

Please request a pickup by 10 am This allows us to ensure speedy turnaround on time-critical harvest analyses.

HOURS Monday – Friday: 7am – 10pm

DROPBOX IS LOCKED FO R T H E P I N C O D E , C A L L U S , O R LO G I N TO YO U R E T S A C C O U N T A N D V I S I T “ C O N TA C T ”

HARVEST HOURS

WEEKEND SCHEDULE

LO D I D RO P B OX E S TAT E C R U S H

F O L L O W U S O N T W I T T E R F O R U P D AT E S : TWITTER.COM/ETSLABS

PICKUP TIME

S AT U R D AY

S U N D AY

AUG 17 – AUG 30

On Call*

On Call*

AUG 31 – SEPT 20

9am - 4pm

On Call*

SEPT 21 – NOV 8

9am - 6pm

9am - 4pm

On Call*

On Call*

Samples are picked up at 10am each weekday, and Saturdays during harvest. Anything left after 10am will be processed with the next pickup.

OPENING HOURS

NOV 9 – NOV 24

2 W. Lockeford Street Lodi, CA

T H E D R O P B O X I S AVA I L A B L E W H E N E S TAT E C R U S H I S O P E N : ( 2 0 9 ) 3 6 8 - 7 5 9 5

* TO SCHEDULE ON-CALL SERVICE, PLEASE C A L L B Y 2 P M O N F R I D AY : ( 7 0 7 ) 9 6 3 - 4 8 0 6

55


HEALDSBURG CALIFORNIA

A F T E R H O U R S D RO P B OX HEALDSBURG

PHONE

(707) 433-7051

Located at our laboratory, on the north side of the building.

COURIER SERVICE Our complimentary courier service is available in Napa, Sonoma, and Mendocino counties every day ETS is open.

Samples left overnight will be processed when we open the following business day.

ADDRESS

REQUEST A PICKUP

190 Foss Creek Circle, Suite G Healdsburg, CA 95448

You can request a courier pickup by logging in to your ETS account, or by calling our lab.

I F YO U N E E D TO S H I P S A M P L E S , P L E A S E S E N D T H E M D I R E C T LY T O O U R S T . H E L E N A LAB: 899 ADAMS STREET, SUITE A ST. HELENA, CA 94574

DEADLINE

Please request a pickup by 10 am This allows us to ensure speedy turnaround on time-critical harvest analyses.

HOURS Monday – Friday: 7am –7pm

DROPBOX IS LOCKED. FO R T H E P I N C O D E , C A L L U S , O R LO G I N TO YO U R E T S A C C O U N T A N D V I S I T “ C O N TA C T ”

HARVEST HOURS

WEEKEND SCHEDULE F O L L O W U S O N T W I T T E R F O R U P D AT E S : TWITTER.COM/ETSLABS

S AT U R D AY

S U N D AY

AUG 17 – AUG 30

On Call*

On Call*

AUG 31 – SEPT 20

9am - 4pm

On Call*

SEPT 21 – NOV 8

9am - 4pm

9am - 4pm

On Call*

On Call*

NOV 9 – NOV 24

* TO SCHEDULE ON-CALL SERVICE, PLEASE C A L L B Y 2 P M O N F R I D AY : ( 7 0 7 ) 9 6 3 - 4 8 0 6

56


PASO ROBLES

CALIFORNIA

A F T E R H O U R S D RO P B OX PA S O R O B L E S

PHONE

(805) 434-9322

Located at our laboratory, next to the front door.

COURIER SERVICE We're excited to launch complimentary courier service this September for our clients in the Central Coast.

Samples left overnight will be processed when we open the following business day.

ADDRESS

REQUEST A PICKUP

3320 Ramada Drive, Suite B Paso Robles, CA 93446

You can request a courier pickup by logging in to your ETS account, or by calling our Paso Robles lab.

I F YO U N E E D TO S H I P S A M P L E S , P L E A S E S E N D T H E M D I R E C T LY T O O U R S T . H E L E N A LAB: 899 ADAMS STREET, SUITE A ST. HELENA, CA 94574

DEADLINE

Please request a pickup by 10 am. This allows us to ensure speedy turnaround on time-critical harvest analyses.

HOURS Monday – Friday: 7am – 7pm

DROPBOX IS LOCKED. FO R T H E P I N C O D E , C A L L U S , O R LO G I N TO YO U R E T S A C C O U N T A N D V I S I T “ C O N TA C T ”

HARVEST HOURS

WEEKEND SCHEDULE

COURIER SERVICE SERVICE AREA

F O L L O W U S O N T W I T T E R F O R U P D AT E S : TWITTER.COM/ETSLABS

S AT U R D AY

S U N D AY

AUG 17 – AUG 30

On Call*

On Call*

AUG 31 – SEPT 20

9am - 4pm

On Call*

SEPT 21 – OCT 31

9am - 4pm

9am - 4pm

On Call*

On Call*

NOV 1 – NOV 24

We're currently picking up samples in the Paso Robles and SLO areas. We will continue to expand as demand grows- follow us on twitter for updates.

* TO SCHEDULE ON-CALL SERVICE, PLEASE C A L L B Y 2 P M O N F R I D AY : ( 7 0 7 ) 9 6 3 - 4 8 0 6

57


NEWBERG OREGON

A F T E R H O U R S D RO P B OX NEWBERG

PHONE

(503) 537-6245

Located at our laboratory, next to the main entrance. Samples left overnight will be processed when we open the following business day.

ADDRESS

COURIER SERVICE Our complimentary courier service is available in Salem, McMinnville, Newberg, and the surrounding areas every day ETS is scheduled to be open.

REQUEST A PICKUP

214 W. Hancock Street Newberg, OR 97132

You can request a courier pickup by logging in to your ETS account, or by calling our lab.

I F YO U N E E D TO S H I P S A M P L E S , P L E A S E S E N D T H E M D I R E C T LY T O O U R S T . H E L E N A LAB: 899 ADAMS STREET, SUITE A ST. HELENA, CA 94574

DEADLINE

Please request a pickup by 10 am This allows us to ensure speedy turnaround on time-critical harvest analyses.

HOURS Monday – Friday: 7am – 7pm

DROPBOX IS LOCKED. FO R T H E P I N C O D E , C A L L U S , O R LO G I N TO YO U R E T S A C C O U N T A N D V I S I T “ C O N TA C T ”

HARVEST HOURS

WEEKEND SCHEDULE

ROS E B U RG D RO P B OX UMPQUA COMMUNITY COLLEGE

F O L L O W U S O N T W I T T E R F O R U P D AT E S : TWITTER.COM/ETSLABS

S AT U R D AY

S U N D AY

AUG 24 – SEPT 13

On Call*

On Call*

SEPT 14 – OCT 31

9am - 4pm

On Call*

On Call*

On Call*

NOV 1 – NOV 24

* TO SCHEDULE ON-CALL SERVICE,PLEASE C A L L B Y 2 P M O N F R I D AY : ( 5 0 3 ) 5 3 7 - 6 2 4 5

58

PICKUP TIME

Samples collected at 11:30 Monday – Thursday. Anything dropped off after 11:30am will be processed with the next pickup.

912 Umpqua College Rd. Roseburg, OR


WALLA WALLA WASHINGTON

PHONE

HOURS

COURIER SERVICE

(509) 524-5182

Monday – Friday: 7am – 7pm HARVEST HOURS

ADDRESS

A F T E R H O U R S D RO P B OX WA L L A WA L L A

3020 E. Isaacs Ave. Walla Walla, WA 99362 I F YO U N E E D TO S H I P S A M P L E S , P L E A S E S E N D T H E M D I R E C T LY T O O U R S T . H E L E N A LAB: 899 ADAMS STREET, SUITE A ST. HELENA, CA 94574

Located at our laboratory, next to the main entrance. Samples left overnight will be processed when we open the following business day.

WEEKEND SCHEDULE

Our complimentary courier service is available in Walla Walla and the surrounding areas every day ETS is scheduled to be open. REQUEST A PICKUP

You can request a courier pickup by logging in to your ETS account, or by calling our lab.

F O L L O W U S O N T W I T T E R F O R U P D AT E S : TWITTER.COM/ETSLABS

S AT U R D AY

S U N D AY

AUG 24 – SEPT 13

On Call*

On Call*

SEPT 14 – OCT 31

9am - 4pm

On Call*

On Call*

On Call*

NOV 1 – NOV 24

* TO SCHEDULE ON-CALL SERVICE,PLEASE C A L L B Y 2 P M O N F R I D AY : ( 5 0 9 ) 5 2 4 - 5 1 8 2

DEADLINE

Please request a pickup by 10 am This allows us to ensure speedy turnaround on time-critical harvest analyses. DROPBOX IS LOCKED. FO R T H E P I N C O D E , C A L L U S , O R LO G I N TO YO U R E T S A C C O U N T A N D V I S I T “ C O N TA C T ”

D R O P B O X L O C AT I O N S PROSSER

R E D M O U N TA I N CO O P E R W I N E CO. AUGUST-JANUARY

RICHLAND CENTRAL INDUSTRIAL SALES

WOODINVILLE W O O D I N V I L L E C U S TO M CRUSH

Dropbox

401 7th St. Prosser, WA

35306 N Sunset Rd. Benton City, WA

2235 Henderson Loop Richland, WA

14030 NE 145th St., Suite B Woodinville, WA

PICKUP TIME: 11AM

PICKUP TIME: 11:30 AM

PICKUP TIME: 12 PM

PICKUP TIME:

PLEASE DROP SAMPLES BY 10:30 A M F O R S A M E - D AY D E L I V E R Y M O N D AY - F R I D AY

PLEASE DROP SAMPLES BY 11 A M F O R S A M E - D AY D E L I V E R Y M O N D AY - F R I D AY

PLEASE DROP SAMPLES BY 11:30 A M F O R S A M E - D AY D E L I V E R Y M O N D AY - F R I D AY

PLEASE DROP SAMPLES BY 2 PM FOR OVERNIGHT SHIPMENT M O N D AY - F R I D AY

59


WWW.ETSLABS.COM S t. H e l e n a C A

|

Healdsburg CA

INFO@ETSLABS.COM |

PA S O R OBLE S C A |

(707) 963-4806

NEWBERG OR

|

Wall a Wall a WA


Turn static files into dynamic content formats.

Create a flipbook
Issuu converts static files into: digital portfolios, online yearbooks, online catalogs, digital photo albums and more. Sign up and create your flipbook.