By Tom Mansell, Science Editor
Photo courtesy of Keuka Spring Vineyards
I recently reviewed Sheldrake Point's 2008 Gewürztraminer for my own site, and as I looked over the technical notes, I noticed that winemaker Dave Breeden added acid to the wine. I was curious about that, given the fact that many wines inthe Finger Lakes are supported by their natural acidity. The concept of adding acid in a cool-climate region like this was a bit strange to me.
And, the alcohol content on this wine was listed at 13.6%, pretty high for whites in the region. Of course, I have seen Alsatian Gewürztraminers come in at 16% ABV. I wondered why the Gewürz that was going into these wines had such low acid and high alcohol (and, in turn, why it is harvested at such high Brix).
"Generally speaking, for Gewürz, acid and pH and really even Brix play no role for me in harvest decisions," says Breeden. For him, the most important parameter for ripeness in Gewürztraminer is the development of that classic lychee/rose varietal character. In the case of their 2008, the pH at harvest was 3.7 and Brix was 23.2, both somewhat high for the region and vintage for the average cultivar. The pH in the final bottling was 3.53.
"I'm a pH winemaker," he said in his characteristic blunt-yet-endearing tone. None of his wines go out the door with a pH above 3.59. The primary reason for this is microbial stability. In the case of Gewürz, acid additions can contribute to balance as well, but the focus is on producing wines able to age, and pH plays an important role in that stability.
So why leave the grapes to hang so long? Fry says "If you want good varietal character, you have to wait for it." Again, it seems like varietal character for Gewürz develops late in the game. If the flavor is right when the grapes are tasted, then it's time to harvest.
The harvest decisions of Breeden and Fry (i.e., waiting until varietal
character develops in grapes by taste) are somewhat uniquely
justifiable for Gewürztraminer. In many cases, varietal aromas in wine comes about when it is synthesized and/or released via
fermentation (think about the difference between the aroma of must and
the aroma of wine). In the case of Muscat and Gewürztraminer, though,
the aroma compounds that define the varietal character are present in
the grapes themselves. Such aromas are known as primary aromas.
Another example is Concord, where the wine tastes like the grapes.
A rose is a rose…?
In 1999, researchers at Cornell determined that the signature aroma of lychee/rose in Gewürztraminer comes from a compound called cis-rose oxide (diagram at right). Unsurprisingly, this molecule is also found in rose petals. It is a product of reactions with monoterpenes (generally, floral aromas) like geraniol and citronellol. The late development of rose/lychee might be attributed to terpene concentrations. Perhaps before cis-rose oxide can develop, the terpene precursors need to build up in the berry.
This hypothesis is supported in the literature. In a landmark paper, Wilson and colleagues noted that after véraison, levels of free and bound terpenes continue to build long after the fruit has reached maturity based on sugar and acid levels alone. After prolonged hang time, those levels began to decrease (likely due to degradation). This phase of terpene decrease could be the optimal ripening point for Gewürztraminer as well, since monoterpenes are the building blocks of cis-rose oxide.
Balancing the unique characteristics of Gewürztraminer
David Whiting of Red Newt Cellars produces two high-end single-vineyard Gewürztraminers (reviewed here and here). "Gewürz has a more complex
balance than most white wines… [It] incorporates balances of
sugar/acid/alcohol/phenolics. This four-way balance can be quite
complex and sometimes counter-intuitive."
Many producers, including Whiting, apply some skin contact (the source of said phenolics) to add complexity and bitterness to the wines. Extended skin contact calls for increased microbial stability (again, pH plays a role). Whiting's grapes are also brought in at high Brix (23.5-25) and acid adjustments are often necessary.
As I mentioned on Monday, over the weekend I got the chance to taste Gewürz selected from around the world, and while I found the South African offering interesting, the clear winner for me was the Albert Mann 2007 from Alsace, which was well-balanced with pure Gewürz characteristics. Albert Mann routinely clocks in at or above 13.5% ABV.
Interestingly, Finger Lakes Distilling has recently released a Gewürztraminer grappa full of awesome Gewürz aromas. The grapes for the grappa were harvested around 23.5 Brix.
In regions like the Finger Lakes, Gewürz lives in the shadow of riesling, and it can often be polarizing. Development of its varietal character takes time, and many winemakers wait for it in spite of (or ignoring) Brix and acidity measurements.
Another important take-home message here is that while Brix and acidity are parameters that can help gauge the ripeness of fruit, individual cultivars can have idiosyncrasies that numbers alone can't account for.
As a final thought, my research into Gewürztraminer has caused me to wonder if there are regions where development of cis-rose oxide and Brix/TA coincide better. Can this be considered analogous to warmer regions where phenolic maturity lags behind sugar and acid numbers? Perhaps there is a climate, somewhere, where all the conditions are right for Gewürztraminer aromas and flavors to come into balance at the same time.
For now, though, I'll enjoy the Gewürz that's being made in New York by winemakers who understand and are willing to tackle the unique challenges its ripening profile and winemaking presents.
Ong and Acree, “Similarities in the Aroma Chemistry of Gewürztraminer Variety Wines and Lychee (Litchi chinesis Sonn.) Fruit“, J. Ag. Food Chem., 1999.
Wilson et al., "Changes in free and glycosidically bound monoterpenes in developing muscat grapes", J. Ag. Food. Chem., 1984.
Luan et al., "Metabolism of geraniol in grape berry mesocarp of Vitis vinifera L. cv. Scheurebe: demonstration of stereoselective reduction, E/Z-isomerization, oxidation and glycosylation", Phytochemistry, 2005.
No umlauts were harmed in the writing of this article.