Ripeness, part 3: solutions to over-ripeness and high alcohol

Part 3 of 3 in a series focusing on the thorny issue of ripeness

Some winegrowers pick late because they have to; picking any earlier would result in unripe characters, chiefly greenness, in their wines. They would also argue that picking earlier wouldn’t give them the flavour profile that they seek. But there’s a feeling that some growers are picking almost perversely late and could very well pick earlier if they chose to, thus solving some of the problems of high alcohol very simply. Also, there’s the question whether all greenness is to be feared in red wines. Classic Bordeaux wines in the past often had a marked green component in their youth and then evolved gracefully; they also commonly had alcohol levels of 12 percent. Classic Rioja and Chianti wines have also shown noticeable green characters in their youth. There is good greenness, some argue, and bad greenness. 

“Viticultural practice has been revolutionized in the New World and sharpened up in the Old World,” says Brian Croser. “We now have very efficient solar panels in our vineyards (vertical canopies with good bud spacing, improved clones and rootstocks, good soil nutrition and water management), all leading to the highly desirable result of more rapid and complete ripening with more anabolic spillover into flavour and colour and better-retained acids.” Croser’s point is that improved viticulture should lead to earlier picking dates, but these are rarely practiced. “A bit like the cork/Stelvin issue [the conflict between supporters of natural cork and screwcaps], there is an element of peer pressure, herd instinct, and browbeating that bullies winemakers to avoid early picking despite the fact that our best vineyards are now set up to facilitate this by achieving early ripeness,” he laments.

“There is some feeling that, in general, New World wines appear to attain the holy grail of physiological maturity at higher levels of potential alcohol relative to Old World wines,” says Randall Grahm.

“The reasons for that I would suggest might be the following (with appropriate remediation proposed):

(1) There is first the question of the length of daylight hours during the growing season, which is dependent on the latitude, i.e. the higher the latitude, the larger the swing in duration of sunlight from the mid-season to the late season. Maybe this triggers some sort of hormonal process in the plant relative to its ripening pattern, and gives the Old World grapes a little goose to get on with it. Alas, nothing to be done in the New World short of an Dr Evil like correction to the earth’s rotational axis.  

(2) In the New World, grapes are generally grown in areas that are strictly speaking warmer than they need to be to fully ripen the fruit (especially if they are cropped at much higher yield levels than one would typically find in Europe, as is generally the case). They are grown this way because they can be—our extraordinarily long growing season allows growers to imagine at least that they can hang these absurdly high crop levels. Growing grapes in cooler climates with restricted yields would certainly help bring the vines into better balance.

(3) Style of farming, especially the utilization of drip irrigation tends to disfavour deeper rooted vines, and this generally leads to problems of dehydration in warm years. This is especially problematic in varieties such as Syrah and Zinfandel. Dry farmed vines, especially older ones, seem to be in just better balance: they throw more appropriate crops and ripen them surely and evenly.

(4) Size of vines. It is generally our wont to create larger vines, more widely spaced than our European counterparts. These bigger vines have more carbohydrate reserves stored in them during the winter and generally grow like crazy in the spring. Perhaps it takes longer for the vine to get the hormonal message that it is time to stop growing and get on with the business of ripening its fruit. (Though it could be argued that this phenomenon may well be counterbalanced by the European phenomenon of summer rain, which can also disrupt the ripening program.) There is the feeling among some people that the larger vines, whether owing to the greater distance between the roots and the trunks and/or the proportionately larger amount of fruit on an individual vine, tend to dilute the mineral concentration of new world wines (which in itself doesn’t bear on the ripening curve, though it may make the wines seem to taste more alcoholic).  But my belief is that smaller vines are more efficient machines, especially with devigorating rootstock, which tend to have a shorter vegetative cycle, i.e. give the plants the signal that it’s time to start the ripening.

Brian Croser highlights some of the viticultural steps that he feels can be taken to counter excessive sugar levels at full ripeness:

“(1) It is necessary to ensure the leaf to fruit ratio is adequate but not excessive (0.7 m2/kg of fruit). Many of the modern vertical canopy vineyards have much too much solar power for the crop load exacerbated by crop thinning (2–3 m2/kg fruit). Nor does just raising the crop level to achieve the lower ratio achieve the right result. Small vines with small crop loads is the answer. I have chosen canopy heights of 500 to 700 mm, not 1 metre plus, with the fruiting wire close to the ground at 0.5 m, not 1 metre plus.

(2) Ensure the daylight photosynthetic and net sugar accumulation time is matched by a nighttime anabolic phase ensuring optimum conversion of sugar to colour and flavour. Low day/night differentials, low vines receiving ground warmth all night is one way to go (Fleurieu Peninsula); another is to allow daytime high temperature to limit the photosynthetic duration and efficiency and to use the cooler but still physiologically appropriate temperatures at night to allow optimum anabolism. Again, small low vines are the answer (Piccadilly Valley). Piccadilly and Fleurieu have the same heat summation about 1170 degree days C and in each case the need is to optimise anabolic phase at night. Sugar accumulation is slower at Piccadilly because of daytime temperatures and picking is later than Fleurieu.

(3) The ripening signals (abscisic acid) are really important to make the anabolic conversion phase go and that relates to the unmanageable average air and soil temperature but the manipulable irrigation regime (partial root zone drying). In the Piccadilly and Fleurieu nature has to take charge of that process but I ensure irrigation is minimized and timed not to interfere with the autumn phase. This is harder on the Fleurieu than Piccadilly. Site slope and soil porosity have to do the work on the Fleurieu to create the right drought stress levels to get the juices running.”

Alcohol levels have been rising for a number of decades, although the figures don’t tell us specifically about the situation with fine wines. The Australian Wine Research Institute published data from their extensive analyses showing average alcohols across all Australian wines sampled rising from 12.4% to 14.2% over the 20 years from 1984 to 2004. And how many of us have had lovely aged Aussie wines from the 1980s and early 1990s with alcohol levels at 13 or below? Julian Alston and colleagues have studied the global rise in alcohol levels. Because label information isn’t accurate, they used a data set from Ontario’s (Canada) Liquor Control Board (LCBO). The LCBO, a monopoly responsible for the sale of all wines in retail shops in the province, test the actual alcohol level of all wines they sell. They noted the discrepancy on declared levels (most wines under-report, but within the legal margins), and also the rise over time, from 12.7% average in 1992 to 13.8% in 2009.   

It is not unusual to find alcohol levels in excess of 14.5% on fine wines from across the world – even Bordeaux in riper years – and relatively rare to find top red wines with alcohols below 13.5%. This is quite a change over just a few decades, and the situation could be worse than this seems, because of the routine under-reporting of alcohol levels by wineries. California allows wineries 1.5% latitude in deviation from label alcohol levels below 14% and 1% latitude above; Australia allows 1.5% across the board; and the European Union allows 0.5%. But is alcohol in itself a bad thing? It has a significant flavour impact, adding sweetness and body, and also heat at higher levels, as well as masking fruity aromas once it gets to 14% and beyond. But the main reason for concern is that it is an indicator of late picking, and while it is possible to have a high alcohol red that still has fruit freshness (the 2010 Angélus would be an example, which is fresh and structured despite having 15.5% alcohol, with a pH of 3.57), it usually means that the wine will have a sweet, jammy, dead-fruit flavour.  


Thus there are a number of steps that can be taken in the vineyard to help synchronize sugar and phenolic ripeness. Should all these fail, technologies now exist to correct high alcohol levels in the winery.

First, we should mention the old trick of adding water to must. Until fairly recently it used to be illegal, but it has been very common in warm climate regions. In California it is now effectively legal to add it, and since 2017 in Australia water is allowed to be added to high sugar musts, but only to bring them back to a potential alcohol level of 13.5%. In most other countries it is still illegal, although some water addition is permitted in the process of adding processing aids and in pumping from tank to tank. Adding tap water can cause problems in the wine, first through the presence of chlorine in tap water (as opposed to groundwater), which can cause fermentation problems, and also because of the risk of geosmin contamination (this earthy-smelling compound is often found in water, especially in warm summer conditions where algal blooms are common).

Three legal technologies are used to reduce alcohol. The first is reverse osmosis, which relies on a technology called cross-flow filtration. A portion of the wine is put through a cross-flow filter, which acts a bit like kidney tubules in that the liquid being filtered isn’t forced through a membrane but instead runs through a tube under pressure, the walls of which are made of the filtration membrane. The advantage of this technique is that the flow of liquid keeps the membrane pores from clogging. However, the surface area of membrane required is enormous. This large surface area is achieved by using a column consisting of numerous small tubes. As the wine passes through, a mixture chiefly composed of water, acetic acid, and alcohol is separated from it. The alcohol is then removed by distillation. The water can then be recombined with the wine to produce a lower-alcohol wine that can be blended back into the original larger batch to produce wine with the desired alcoholic strength. The second technique, called Memstar, is similar, but relies solely on membranes, with a second membrane step for separating the water and alcohol fraction rather than a distilling column. The cost of a reverse osmosis set-up starts at around $50 000, whereas a Memstar starts at around $100 000. Both these techniques require the wine to be passed through at pressure, and there is a little loss of flavor. But they are the only option for estate wines that are legally not allowed to leave the winery premises. 

The third technique, the spinning cone, achieves the same ends by rather different means. Originally invented for making heavy water for the nuclear industry, but commercialized for wine by the U.S. company Conetech, the spinning cone column contains around forty upside-down cones, of which half are fixed and half spin. In a vacuum environment, the cones spin the wine into thin liquid films, and a cool vapour rises off the wine, carrying the volatiles from the liquid. In the first pass, the ultralight component consisting of the delicate flavours and aromas is carried off and condensed. This is known as the essence, and it is saved to be recombined with the wine later. The second pass takes off as much alcohol as you want to remove. Theoretically, you could then recombine the remaining low-alcohol wine with the essence and the alcohol and end up with the same wine you started with.

Currently, Conetech treats wines from around six hundred clients worldwide, and it treats around 6 million gallons. However, because it treats only a small proportion of each wine—around 10 percent—that is then blended back, around 50 million gallons of wine have the alcohol level reduced in this way. As well as a plant in California, Conetech also has plants in Chile, South Africa, and Spain. The cost of the spinning cone machine is around US$1 million.

The rules are potentially a problem for this sort of technology. In the United States, spinning cones have been authorized for alcohol reduction, but in Europe this technique used to be allowed only on an experimental basis. This meant that you were allowed to treat 50,000 hectolitres, but the wine couldn’t leave the country of origin. But in November 2008 the EU regulations changed to make it legal to remove up to two degrees of alcohol where specific local appellation laws permitted this (in many appellations in France, for example, this is still not permitted).

Tasting a series of samples of the same wine with the alcohol reduced, both by the spinning cone and by reverse osmosis, reveals how much of an impact the presence of alcohol has on the perception of other wine components.  “Alcohol is a masking agent,” says Conetech’s ex-head winemaker Scott Burr, “so taking it away reveals what’s there. It also adds sweetness to the palate.”

How does a winemaker decide how much alcohol should be reduced? “A big client who we’ve been working with a long time might specify they want their wine at 13.85 percent,” says Burr. “A smaller client might say they have a Zinfandel at 16.8 percent alcohol, so we do a run, take the 4 percent component and blend wines at a bunch of different alcohol levels,” he says. “I don’t tell the client what the alcohol level should be.” In regard to which wines are more successful, Burr says that there is not a curve; instead, there are sweet spots. “There are all kinds of different ones. The more oak, the bigger the variance.”

Clark Smith, a California winemaker whose previous company Vinovation was the leading practitioner of reverse osmosis in the United States, reckons that 45 percent of premium California wines are alcohol adjusted, either by reverse osmosis or by the spinning cone. Alcohol reduction is becoming a widely adopted tool throughout the winegrowing world, but it is not without controversy, simply because it’s seen as a rather artificial technique that subjects a portion of the wine to fairly dramatic physical forces.

Clearly, such interventionist strategies don’t fit well with the concept of natural wine. But there is an interesting, almost philosophical question here. If a grower is stuck in a position where he or she is forced to choose between flavour development and sensible alcohol levels, whatever he or she does in the vineyard, might this justify the use of alcohol-reduction technologies as a last resort? They do sound horridly manipulative, and most fine wine producers prefer to do as little winemaking intervention as possible. But the tantalizing possibility remains that wines that speak more eloquently of their origins—the vineyard site or terroir they came from—could be realized by the use of these tools. We know it sounds like heresy, but could alcohol-reduction techniques in the wine cellar assist in the expression of terroir that traditionally minded winegrowers seek? Could reverse osmosis or the spinning cone have an important role to play in fine wine production, compensating for the effects of global warming?

The argument would be that technologies such as this are just tools, and tools are morally neutral—it is how they are used that matters. A winegrower who genuinely wants to make a wine that expresses the vineyard site optimally, but finds the wine’s ability to do this hindered by the masking, sweetening effect of high alcohol, could use such a tool to produce a better wine that has more of a sense of place to it (in theory, at least). 

“In the past, I confess to having used some technological methods (spinning cone and reverse osmosis) to remove alcohol from our wines,” admits Randall Grahm, “but I am now quite opposed to the practice (and of adding water as well—known as ‘Jesus Units’ in the trade). If you are doing the right work in the vineyard, you should not be compelled to resort to these extreme solutions.”

He continues: “In the winery, the most practical solution that we have found to keeping alcoholic degree in check has been the use of open-top fermenters, warm fermentation temperatures, and especially the use of indigenous yeasts. The ‘wild’ yeasts have at least for us been absolutely brilliant; they tend to give us a much longer, more even fermentation (and also don’t stick). But what is particularly cool is that the fermentation process is far less efficient, owing to the successive populations and obscure yeast wars going on between species. They’re essentially converting the sugar in the grapes into biomass rather than alcohol and one ends up with very low sugar to alcohol conversion levels, in some cases approximately 0.51–0.52 (Brix to percentage alcohol).”

This last point is quite controversial. The conversion level Grahm quotes is contested by some yeast researchers we contacted; they don’t think that the yeasts would convert sugar to biomass, and they add that even in an indigenous fermentation, most of the fermentation will be by means of Saccharomyces cerevisiae species, which will have a higher conversion factor, about the same as that of selected yeast species.

One of the current objectives of yeast research is to produce wine yeast strains that are less efficient at converting sugar to alcohol. Such strains already exist, but the two key problems have been finding non-GM yeasts that have these properties, and finding strains that use the sugar to produce alternative fermentation products that have desirable sensory properties. GM strains of S. cerevisiae can produce wines that have 2 percent lower alcohol, but they couldn’t be used in wine. The greatest variation in alcohol levels that has been demonstrated with natural fermentations with non-GM S. cerevisiae is 0.2 percent, which is much less than the variation achieved by modifying the fermenter design. Still, it seems a bit like a sticking-plaster solution to a more fundamental problem, and the major focus of those looking to produce wines with more sensible alcohol levels should be on viticultural practice.

With the world of wine it seems that there are competing aesthetic systems. There’s the classic world of fine wine, based upon the great wines of Europe. To the old school wine trade of 40 years ago, the modern 15% Napa Valley Cabernets would be wines that simply wouldn’t fit into this aesthetic system. They are entirely outside it. Parker, with his confidence in his own palate, was prepared to operate outside this system. He helped usher in a new aesthetic system of fine wine that was overlapping with the old system, but which embraced styles of wines that were quite new, and included them alongside the established classics. Perhaps the way for Parker was made easier by the emergence of New World wines, and Spurrier’s famous 1976 Judgement of Paris tasting, which began to question the supposed supremacy of the old world standards. Now it seems, a new aesthetic system is emerging, driven by a new generation of wine professionals who find the ripe, alcoholic style beloved by the influential critics to be distasteful and even ludicrous. This new aesthetic values stylistic diversity and lesser known varieties and regions, alongside the best of the classics, with the proviso that the wines express a sense of place. Picking late is an enemy of this terroir expression. What we are witnessing with the contentious nature of IPOB is explainable as a clash between competing aesthetic systems. They are different, and perhaps in this case more different than the aesthetic systems that collided with the emergence of Parker and other influential American critics back in the 1980s and 1990s. Thus we could argue that this move away from excessive ripeness isn’t just a pendulum swing, or a correction. It’s a proper revolution; a regime change.

I will leave the final words to Cain’s Chris Howell. “Taste is not only individual, it is also cultural. New wine drinkers, as we were all once, are always asking, what do we like, what should we like, what is good?’ Howell concludes, ‘The answer may be personal, but it is also a shared conversation. Our wine culture is in constant evolution.” Long may this important discussion continue.

Ripeness PART 1, PART 2, PART3