Learning to Sniff Out Corked Wine

Yes, I'd like to think my motives were scientific, but I suspect the real reason I wanted to challenge a dog to a wine sniff-off was because of the time I got nailed by the airport customs beagle. I was just back from a wine trip to Spain when the beagle, looking spiffy in a green jacket, made a beeline for my suitcase and woofed at it. That was the end of my carefully wrapped package of jamón ibérico.

"You can't bring that into the country," said the beagle's handler, an unsmiling woman in a uniform.

"Can I give it to the dog?"

"Of course not."

The beagle looked at me smugly, as if to say, "I am an official U.S. Customs and Border Protection Ag Detector Dog. And you, you ham smuggler, are a bonehead."

So maybe it was a bit of getting my own back, because a few years later, here I was at Sojourn Cellars in Sonoma, California, ready for a nose-to-nose showdown with Ziggy, a five-year-old fox-red Labrador retriever who'd been trained since birth to sniff out TCA.

It might be useful to explain what TCA is. It is 2,4,6- trichloroanisole, a chemical substance that makes winemakers grit their teeth and mutter expletives, because it is responsible for what's known as "cork taint." Cork taint is a big problem for anyone who makes or drinks wine. Whether the wine is a $200 Bordeaux or a $5 Shiraz, that 25-cent cork stuck in the neck of the bottle can, if contaminated with TCA, render it worthless. What's even more frustrating is that many people who buy wine are uncertain how to tell whether a bottle is contaminated, so most end up drinking it anyway, then sitting back and thinking, Well, that wasn't very good.

So here's what you need to know. First, when someone says a wine is "corked" or "corky," they mean that the cork, and thus the wine, has been contaminated with TCA. (They don't mean that there are stray bits of cork floating around in the liquid.) Depending on the amount of TCA present, the wine will range from smelling like nothing at all—low levels of TCA will kill a wine's aroma and flavor, even if the scent of the TCA itself isn't discernible—to smelling like an old, damp cardboard shoe box removed from under a pile of moldy newspapers.

TCA is incredibly potent. One ounce of pure TCA would be enough to noticeably contaminate about 10 billion bottles of wine, or more than five times California's annual wine production. Luckily, TCA tends to occur in amounts measured in parts per trillion, rather than ounces. But even at those microscopic levels, its presence can be damaging. Almost anyone can pick up the musty, damp scent of TCA at about 10 parts per trillion. Swirl the wine in a glass, sniff it, and if the aroma evokes someone's grandfather's basement, that bottle is corked. Experienced wine tasters can sometimes detect TCA in concentrations as low as one part per trillion.

A certain percentage of wine corks (and, less frequently, barrels or wooden pallets—also, rarely, entire wineries) end up contaminated with TCA. The chemistry of why this occurs is complicated. Essentially what happens is that phenols in cork-tree bark, when exposed to chlorine compounds (for various reasons), react with fungi or mold in the bark to produce TCA. How often this happens is much debated. The cork industry finds that TCA contaminates a very small number of corks, less than one percent (improved quality controls have helped in recent years). But winemakers tend to find the prevalence to be much higher, from three to five percent. What no one disagrees on is that some percentage of corks are tainted.

If you think about it, the fact that people put up with this kind of contamination is bizarre. What would happen, in contrast, if five percent of all the milk sold in grocery stores was spoiled? Of course, the fact that TCA's effects can be subtle, and that people generally aren't as familiar with wine as they are with milk, may explain why there isn't a public outcry against this problem.

But there has been a reaction within the wine industry. All this TCA trauma has led to the rise of alternative seals for bottles—screw caps, glass stoppers, plastic corks, crown caps and the mysteriously named Zork, a kind of plastic stopper, among others. It's also, to some degree, a reason why someone would train a dog to sniff out TCA.

Of course, I didn't fly out to Sonoma to meet Ziggy solely because of my run-in with the airport customs beagle. It was more a way to make a point, which is that any creature with a sense of smell can learn to identify TCA. My wife, Cecily, for instance—a charming woman who would probably look alarmed if anyone accused her of being a wine expert—can spot cork taint with the best of them. As far as I can tell, this is simply because she lives with an odd man who often shoves glasses of wine at her and says, "I could swear this thing's corked. What do you think?"

In Sonoma, I drove to Sojourn Cellars to meet Ziggy's owner, Craig Haserot, who makes very good Pinot Noir when he's not entertaining strange requests from wine writers. He told me that Ziggy had previously been owned by a friend who supplies oak staves, chips and dust to the wine industry; this friend was the one who'd decided that a dog trained to sniff out TCA would be helpful around a drying facility, as a backup to chemical testing. This wasn't a far-fetched idea: Humans have about 12 million olfactory receptors, whereas dogs have about 220 million, and when was the last time you heard about someone using a human to sniff out bombs? Unfortunately, it turned out that the friend traveled internationally too much to care for a dog, and so he bequeathed Ziggy to Haserot.

"She was in a really gnarly, hard-core training program," Haserot said. Ziggy came over and sat down on my foot, looking neither gnarly nor hard-core, but cute and sweet. "She came out of the kennel knowing how to find TCA, but nothing else. She didn't even know how to play." Ziggy looked up at me plaintively. "She wants her b-a-l-l," Haserot explained. Apparently, playing wasn't a problem anymore.

Well, ball time later for you, hound, I thought. Haserot had cut up a barrel stave earlier and dipped one piece in a very diluted TCA solution, somewhere under two parts per trillion. We hid them, and Haserot said, "Ziggy! Find it!" Ziggy zigged and zagged—she searches in a back-and-forth pattern, hence her name. She found the non-TCA stave, ignored it, found the TCA stave, grabbed it in her teeth, and brought it to Haserot, who gave her a dog treat.

Then it was my turn. I wasn't going to roam around on all fours—I had to maintain a certain amount of dignity here—but I did attempt to differentiate between the pieces of wood with my eyes closed. Now, my ability to pick up TCA is pretty good for a human. (I had my TCA threshold tested at a Napa company called Vinquiry for $75, something a lot of wineries have their staff do, and apparently I can pick it up at levels of one part per trillion; I even got a nifty certificate to that effect.) But when it comes to detecting trace amounts of TCA on oak barrel staves, compared to Ziggy, I'm an olfactory dolt. I could point out that chunks of oak don't smell like glasses full of wine; I could also argue that the presence of dog slobber adds an additional aromatic note to oak that might, in my case, possibly have obscured the faint scent of TCA. On the other hand, I could also just admit that I lost. I'm a good sport. I knelt down. "Good job, Ziggy. Shake."

Ziggy lifted a paw. We shook. And, unlike that beagle in the airport, she didn't gloat at all.