Winemaking Laboratory

Chapter I · Vine and terroir

Ripening dynamics

Sugar, acids, pH and phenolics from véraison to harvest — how berry composition changes under the influence of heat, and when the optimal harvest window opens.

Site and variety

Wine style and criteria

Semi-empirical model: sugar — logistic function of accumulated heat; malic acid — Q₁₀ kinetics; anthocyanins — with a temperature optimum. Dates are accurate to ±1–2 weeks; the final decision is made from field samples.

Véraison—
Technological ripeness—
Phenolic ripeness—
Harvest window—

Must composition

Sugar—°Brix
Oechsle—°Oe
Baumé—°Bé
Potential alcohol—% vol
Titratable acidity—g/L
pH— 
Sugar / acidity—B / TA
Ripeness index—pH² · B

Four temperature scenarios

Site, variety and style criteria are taken from the "Ripening curves" mode:

Scenarios are added to a normal year, not to the anomaly chosen in the first mode. +3 °C roughly corresponds to high-emission climate projections for the region in the second half of the century.

ScenarioVéraisonMin. Brix reachedTA thenpH thenPhenolic thenHarvest window

Field samples

DateBrixTApH

TA in g/L as tartaric acid. Blank TA or pH values are allowed — they are simply left out of the corresponding trend. A series of Saperavi samples is pre-filled as an example.

Sugar accumulation rate—°Brix / week
Date of target sugar—
TA on that day—g/L (forecast)
pH on that day—forecast
DateBrix°Oe°BéPA %TApHB/TApH²·B

Intuition

Over the summer a grape berry lives two completely different lives. Before véraison it is green, hard and sour — like a leaf, it photosynthesises and accumulates acids. At véraison everything changes within a few days: the berry softens, the skin changes colour, sugar flows in rapidly through the phloem, and the acids, by contrast, are "burned". Ripeness is the moment when these processes reach the balance we want.

The problem is that "ripe" cannot be measured with a single number. Sugar content determines the alcohol of the future wine; acidity and pH determine its freshness, microbiological stability and the effectiveness of SO₂; phenolics determine colour, tannin structure and — for qvevri wine — its entire character; aroma compounds determine varietal identity. These processes run at different speeds and respond differently to heat. That is why, in the same vineyard, sugar may "ripen" well before phenolics in a hot year, while in a cool year the opposite happens — phenolics are ready but sugar is still insufficient.

Technological ripeness is the moment when sugar and acidity match the desired wine style. Phenolic ripeness is when skin anthocyanins are close to their maximum and seed tannins have "matured" (become less harsh and bitter). The optimal harvest window is the overlap of the two — and the calculator on this page is designed to find exactly that window.

History

For centuries the harvest date was decided by taste, berry colour and the colour of the seeds — the Georgian rtveli was traditionally decided by the long experience of the family and the village. Quantitative measurement began in the 19th century: in the 1830s Ferdinand Oechsle built a hydrometer that measured must density (Oechsle degrees are still used in Germany and Switzerland); Adolf Brix created a scale for sugar solutions that is today the most widely used unit in the world; in French practice the Baumé scale took hold.

In the second half of the 20th century the Australian physiologist Bryan Coombe described the double sigmoid curve of berry growth and véraison as a sharp physiological switch. Research in the 1960s–70s by Hans Ruffner and others showed that the breakdown of malic acid in the berry is a respiratory process and depends on temperature — which explains why grapes from warm regions are less acidic. In the 1990s, at the University of Bordeaux, Yves Glories developed a method for analysing phenolic ripeness (extraction of anthocyanins at pH 1 and pH 3.2), which made "phenolic ripeness" measurable in the laboratory.

Stages of berry development

From flowering to harvest, berry growth follows a double sigmoid curve. The first stage (roughly 45–60 days after flowering) is cell division and expansion; the berry is green and actively accumulates tartaric and malic acids, as well as hydroxycinnamic acids and seed tannins. The second stage is the "lag phase", a slowdown of a few days to a few weeks, during which the seed forms and berry size hardly changes.

The third stage begins with véraison. At this point the xylem water supply drops sharply and the berry is fed mainly through the phloem, which brings in sucrose; invertase splits it into glucose and fructose in a ratio of roughly 1:1. The berry volume almost doubles, sugar content rises from 4–5 Brix to 20–25 Brix, and in red varieties anthocyanins accumulate in the skin. Late in ripening, in some varieties the phloem flow stops and sugar continues to rise only through dehydration (berry shrivel).

Mathematics

The model on this page is semi-empirical: it uses known regularities of the biochemical processes, with parameters fitted to typical values from the literature. Véraison occurs when accumulated degree days reach about 62% of the variety's ripening requirement: $G_v = 0.62\,G_{th}$.

Sugar accumulation after véraison is described by a logistic (S-shaped) function whose argument is accumulated heat rather than calendar time:

$$B(G) = B_0 + (B_{max} - B_0)\,\frac{L(G) - L(G_v)}{1 - L(G_v)}, \qquad L(G) = \frac{1}{1 + e^{-k(G - G_{50})}}$$

where $B_0 \approx 5$ Brix (the moment of véraison), $B_{max}$ is the variety's sugar potential (20–27 Brix), $G_{50}$ is the midpoint of accumulation, and $k$ is chosen so that at the variety's heat requirement ($G_{th}$) the logistic function reaches 90% of its maximum.

The breakdown of malic acid is a first-order reaction whose rate depends on temperature according to the $Q_{10}$ law:

$$\frac{dM}{dt} = -r_{20}\,Q_{10}^{(T - 20)/10}\,M, \qquad Q_{10} \approx 2$$

$Q_{10} = 2$ means that every 10 °C of warming doubles the breakdown rate. This is why grapes lose acidity much faster in a hot year — even at the same sugar level. Tartaric acid is metabolically stable in the berry and its concentration is only diluted by berry growth. Titratable acidity as tartaric acid equivalent: $TA = T + 1.119\,M$ (the coefficient 1.119 = 150/134 is the ratio of the molar masses of tartaric and malic acid).

The empirical relationship between pH and titratable acidity (for an average potassium content): $pH \approx 4.39 - 1.144\,\log_{10} TA$. In reality potassium also strongly affects pH — on potassium-rich soils pH is higher at the same TA.

Anthocyanin synthesis is a process with a temperature optimum (around a mean temperature of 20 °C); it is inhibited when the daily maximum exceeds 30 °C and enhanced by a large diurnal temperature range:

$$\frac{dA}{dt} = k_A\,f(T)\,g(\Delta T)\,(A_{max} - A), \qquad f(T) = e^{-\left(\frac{\bar{T} - 20}{8}\right)^2}$$

Conversion between must density units: $SG = 1 + \frac{B}{258.6 - 0.8796\,B}$; Oechsle: $°Oe = 1000\,(SG - 1)$; Baumé: $°Bé = 145\,(1 - 1/SG)$. Potential alcohol: by EU convention 16.83 g/L of sugar gives 1% vol of alcohol; the sugar concentration of the must is approximately $S \approx 9\,B\,SG$ g/L (about 90% of soluble solids are fermentable sugars).

Key quantities

QuantitySymbolUnitTypical value at harvest
Sugar content (Brix)B°Brix (g/100 g)19 – 25
Oechsle degree°Oe—80 – 110
Baumé degree°Bé—10.5 – 14
Potential alcoholPA% vol11 – 14.5
Titratable acidityTAg/L (tartaric)5 – 9
pHpH—3.1 – 3.7
Malic acidMg/L1 – 5
Tartaric acidTg/L4 – 7
Anthocyanins (red varieties)Amg/kg berries500 – 2000
Sugar/acid ratioB/TA—2.5 – 4
Ripeness index$pH^2 \cdot B$—200 (white) – 260 (red)

Harvest indices

The ratio of sugar to acidity (B/TA) is the oldest index: it rises with ripeness and combines the increase in sugar and the decrease in acid in a single number. Its weakness is that very different musts can have the same ratio — for example, 20 Brix / 7 g/L and 24 Brix / 8.4 g/L.

A more informative index is $pH^2 \cdot B$ (Coombe and co-authors, 1980), because pH reflects the chemical behaviour of wine better than TA. As a general guide: about 200 for white wine, about 260 for red. These figures are a guideline, not a rule — the specific style, variety and winemaking change them.

Wine styleBrixTA (g/L)pHPhenolic ripeness
Sparkling wine base17 – 19.5≥ 8≤ 3.25not critical
Dry white (European)20 – 22.5≥ 6≤ 3.4not critical
Qvevri amber21 – 23.5≥ 5≤ 3.6seed maturity ≥ 70%
Dry red22 – 25≥ 5≤ 3.7≥ 80%
Naturally semi-sweet24 – 28≥ 6≤ 3.6≥ 80%

Monitoring ripeness in the field

Any model is only a forecast — the final decision is made by field sampling. Standard practice: from 2–3 weeks after véraison, once a week, and twice a week as harvest approaches, about 200 berries are collected from a uniform plot — from different rows, both sides of the vine, and the top, middle and bottom of the cluster. Whole-cluster sampling (10–20 clusters) is more representative but labour-intensive. The sample is pressed and the must is measured for sugar (with a refractometer or hydrometer), TA (by titration with NaOH to pH 8.2) and pH.

In Georgian practice the refractometer reading is often referred to as "sugar percent" — this is practically the Brix degree. Refractometer readings need temperature correction (modern instruments do this automatically).

Phenolic ripeness is assessed in the laboratory by the Glories method, and in the field by sensory analysis: seed colour (from green to brown), seed taste (bitter and harsh, or nutty), the feel of skin tannins (coarse, drying or velvety) and how easily the skin releases colour when rubbed between the fingers. For qvevri wine seed maturity is decisive — seeds and skins are in contact with the wine for months, and an unripe seed will give coarse, bitter tannins.

Connections

This page directly continues the terroir and climate page: the same regional temperatures and elevation correction determine the heat accumulation on which the sugar model is built, and the same variety requirements are taken from the variety atlas. The second chapter (grape chemistry and harvest) starts where this page ends — converting Brix to potential alcohol, acid adjustment and titration technique are discussed there in detail.

In the third chapter (alcoholic fermentation) the sugar content measured at harvest is the initial condition for fermentation kinetics. In the fourth chapter pH determines the proportion of molecular SO₂ — between pH 3.2 and 3.7 the free SO₂ needed for protection changes roughly threefold. In the fifth chapter (the qvevri method) phenolic ripeness becomes the main criterion, because long qvevri maceration extracts everything in the skins and seeds — the good and the bad alike.

Application — rtveli in Georgia

In Kakheti the rtveli (harvest) traditionally runs from mid-September to early October: early white varieties and sparkling wine base are picked first, Rkatsiteli and Mtsvane Kakhuri next, and Saperavi — which ripens late and keeps high acidity for a long time — often last. Saperavi's high tartaric acid and intense anthocyanins (it is a teinturier variety, with coloured pulp as well as skin) allow the producer to wait for full phenolic ripeness without a catastrophic loss of acidity.

In Imereti, with warmer nights and a humid autumn, Tsitska and Tsolikouri are picked early and at relatively low sugar — the main risk here is fungal disease rather than over-ripeness. In Racha, Aleksandrouli and Mujuretuli for Khvanchkara are deliberately picked late, at high sugar, because the semi-sweet style requires fermentation to stop naturally while residual sugar is retained.

Climate change makes the "decoupling" of sugar and phenolics the main challenge: in hot years sugar reaches the target level earlier, still in hot August, when anthocyanin synthesis is inhibited and the seeds are still green. The producer faces a choice: harvest early (lower alcohol, but green tannins) or wait (ripe tannins, but high alcohol and low acidity). The calculator's "Vintage scenarios" mode shows this dilemma quantitatively.

Glossary

TermDefinition
VéraisonThe start of the third stage of berry growth: softening, colour change, rapid sugar accumulation
Technological ripenessThe balance of sugar and acidity that matches the desired wine style
Phenolic ripenessAnthocyanins approaching their maximum and the maturing of seed and skin tannins
Must densityThe specific gravity of the must, the main measure of its sugar content (Brix, Oechsle, Baumé)
Titratable acidityTotal acidity measured by titration, expressed as tartaric acid equivalent
Malic acidThe acid that is broken down by respiration during ripening; its decline depends on temperature
Tartaric acidThe main, metabolically stable acid of the grape
AnthocyaninsThe red-purple pigments of the skin, the source of wine colour
Seed maturityThe browning of the seed and the decrease in the harshness of its tannins
Double sigmoid growthThe berry growth curve with two active stages and a slowdown between them
Berry shrivelLate-ripening dehydration, which raises sugar at the expense of volume
Ripeness index$pH^2 \cdot B$ — a combined measure of pH and sugar content

Practical use of the calculations

A Kakhetian winemaker is making a dry red wine from Saperavi in qvevri. At the end of August the field samples show: 15.2 Brix on 20 August, 17.4 on 27 August, 19.3 on 3 September and 20.8 Brix on 10 September, while TA has fallen from 14.5 to 9.3 g/L. The monitoring mode fits a linear trend to these samples: sugar is rising by about 1.9 Brix per week, so the target of 23 Brix will probably be reached around 18 September, by which time TA will be about 7.8 g/L — good acidity for the style. But the ripening-curves mode for the same region and variety shows that in mid-September the phenolic ripeness index is still only about 60% — the seeds are not yet fully mature, and with an 80% phenolic criterion the harvest window opens only in early October. For long qvevri maceration an early harvest is a risk. The winemaker decides to wait about two weeks, at the cost of sugar rising to 24 Brix (about 0.6% more potential alcohol), and in return to obtain ripe seed tannins. Meanwhile he continues sampling twice a week and assessing seed colour and taste, so that the model's forecast is checked against reality.

For beginners — when should we pick?

1. Open the "Calculator" and stay in the "Ripening curves" mode. Choose the region, vineyard elevation and variety — the list of varieties is the same as in the variety atlas.

2. Choose the wine style (for example "Dry red" or "Qvevri amber"). The style automatically fills in the target sugar, minimum acidity, maximum pH and the phenolic ripeness requirement. You can also change these numbers by hand.

3. Press "Calculate". The cards at the top show the date of véraison, the dates of technological and phenolic ripeness and the harvest window — the days on which all conditions are met at the same time.

4. The first chart shows how sugar rises; the green band is the harvest window. The second shows how acidity falls (tartaric acid hardly changes, malic acid burns away quickly) and how pH rises. The third shows colour and seed maturity.

5. If no window appears at all, it means that at this site, with this variety and for this style, the conditions are never met at the same time. The card at the top tells you which condition fails — you can change the style, elevation or variety.

For professionals — the model, scenarios and field data

1. The model is semi-empirical and is a forecasting tool, not a substitute for field analysis. Its main value is relative comparison: how the window shifts 300 metres higher, in a hot year or with another variety. Absolute dates are accurate to ±1–2 weeks and depend on the quality of the climate data and on the correctness of the variety's heat requirement.

2. The "Vintage scenarios" mode compares the same variety and site under four temperature scenarios: a cool year (−1.5 °C), a normal year, a hot year (+1.5 °C) and a climate change scenario (+3 °C). The sugar–acidity trajectory chart shows that in a hot year TA is lower at the same Brix, and the table shows that phenolic ripeness is lower when the target sugar is reached. This is the quantitative picture of sugar–phenolic decoupling.

3. In the "Field monitoring" mode, enter your actual samples (date, Brix, TA, pH). For each sample the tool calculates °Oe, °Bé, potential alcohol, B/TA and $pH^2 \cdot B$; it fits a linear trend to sugar and an exponential decline to acidity over the latest samples, and estimates when the target sugar will be reached and what TA will be on that day.

4. A linear trend works well in the middle of sugar accumulation (roughly 15 to 22 Brix). At the end of ripening accumulation slows (phloem flow decreases), so in the last weeks a linear forecast is optimistic — increase the sampling frequency in this period. A temporary drop in sugar after rain (berries taking up water) is normal and should be allowed for when judging the trend.

5. The relationship between TA and pH depends on potassium. If your measured pH is systematically higher than the model's forecast at the same TA, the plot is rich in potassium — in that case the pH criterion is reached earlier and the window becomes shorter.

1. Why do grapes lose acidity faster in a hot year?

During ripening malic acid is a respiratory substrate; every 10 °C of warming roughly doubles its breakdown rate. Tartaric acid, by contrast, is metabolically stable.

2. What does "decoupling of sugar and phenolics" mean?

Sugar accumulation speeds up with heat, whereas anthocyanin synthesis is inhibited above about 30 °C — so in hot years the two kinds of ripeness drift apart.

3. In Georgian practice, refractometer "sugar percent" corresponds practically to which unit?

°Brix is grams of soluble solids (mostly sugar) per 100 g of solution — that is, a mass percentage.

4. Why is seed maturity especially important for qvevri wine?

Long maceration extracts everything in the skins and seeds. Ripe, browned seeds have less aggressive tannins.