When Bacteria Attack

This post is the final word in a trio depicting the derailing of my final 2010 fermentation - the Anderson Valley fruit.  After three picture perfect fermentations, I was undone by the malodorous stench hanging over my bin.  Fermentation had commenced without my approval.  Here’s how it went down and how it was resolved. The final bludgeoning of harvest hubris was discovering lactic acid in my fermentation bin.  Identifying lactic acid means pretty much one thing: you’ve got lactic acid bacteria.  In wine circles, breeding lactic acid bacteria is not necessarily a bad thing.  These microscopic dudes are responsible for the malo-lactic conversion, also known as “secondary fermentation.”  In my wines, this conversion is not only desirable but actively encouraged…but on my terms.  Like native alcoholic fermentation, native malo-lactic conversions risk courting unwanted strains that not only emit off odors/off flavors but also crap out before the conversion is 100% complete.  Consider for instance an incomplete native malo-lactic conversion.  The bacteria kinda-sorta eat up the residual nutrients and use most-but-not-all of the available malic acid for energy.  Then you sock away your barrels in cold storage until spring, or even summer when you’re ready to bottle.  Meanwhile the cool barrel room upsets their sensitive dispositions, and these fragile Sleeping Beauties slumber soundly until the warm kiss of summer sun awakens them from drowsy stupor.  Your unruly microbes are roused just in time for bottling, where they go on to finish “secondary fermentation” inside the bottle.  Now you have 100 cases of fizzy, turbid, cloudy wine that looks like brown champagne and probably smells even worse.

It goes without saying that I want to control this process as much as Mother Nature lets me wrangle her kin.  We’ve had excellent results using a particular commercial bacterial strain after alcoholic fermentation is complete. It’s freeze dried (like those strawberries from Trader Joe’s), and I sprinkle it right on top of the wine, like fairy dust.  An added bonus is that the dead, decomposing yeast are still releasing their innards into the surrounding wine.  This bathes my rehydrating bacteria in a yummy brew of micronutrients and amino acids provided free of charge by their composting kin.  It’s uber-recycling; one bug’s rotting ribosome is another microbe’s feast.  Of course it all sounds great in a blog post until it fails in practice.

The preliminary lab analysis alerted me to the unwanted infestation.  In a 48-hour juice panel, malic acid should be in the ballpark of 0.5g/100mL, give or take some (depending on ripeness, temperatures, cultivar etc.).  Malic and tartaric acid are the primary grape acids and ride in with the fruit.  They are a natural part of grape juice, made by the plant.  Lactic acid is not.  Recall my juice panel showed 0.179g/100mL malic acid (low) and 0.087g/100mL lactic acid (curious).  Plus my pH had catapulted from pH=3.34 to pH=3.86 while the titratable acid plummeted from 7.88g/L to 4.55g/L.  Basically, I’m paddling a punctured canoe leaking acid all the way to shore.  The dropping acid, rising pH, and unsolicited lactic acid are evidence enough of malo-lactic conversion; now I just needed DNA fingerprints to pinpoint the culprits.

The overpriced “scorpion” panel from ETS Labs (cha-ching!) revealed a mixed bacterial bag.  All bugs fell under the rubric of “lactic acid bacteria,” meaning when they digest stuff (like sugar) they fart out lactic acid as an end product.  In my sample, a handful of Lactobacillus species were cohabitating with Pediococcus - usual stuff.  Like yeast, a number of bacterial critters live on grape berries and hitch a ride to the winery during harvest.  Many succumb to the sulfur dioxide added at the crusher followed by the one-two whammy of low pH (too noxious) and a cold soak (too chilly).  But the species who survive when these measures fall short were the freeloaders assaulting my bin.  Common as they are, their vital stats are surprising.  Take L. plantarum, the “native” most frequently isolated in laboratory samples.  He’ll convert sugar to lactate alright but left to his own devices, he gets more bang for his buck converting sugar to acetate.  And the last thing I need is a vat of vinegar.  Plus he’s been shown to spit out compounds that inhibit yeast.  Now if alcoholic fermentation fails too, you’re left with sweet vinegar… even worse.  How about L. brevis, capable of surviving in 16-20% alcohol?  Usually the “native” hitchhikers die from alcohol poisoning during fermentation, clearing house for a healthy inoculum of commercial bacteria later on.  But 20% alcohol tops even our 2010 Rockpile zin (just kidding- 19.9%). 

Let’s play theoretical now.  Instead of hyperventilating over my lack of control, say I’d let those bacteria ride.  Since I’ll ultimately want the malo-lactic conversion anyway, what’s the harm?  There are two good answers at hand.  The first involves the timing of alcoholic fermentation and malo-lactic conversion, moderating each to ensure the success of the other.  The second is the manipulation of end products.  Most data (UC Davis coursework included) supports malo-lactic conversion either prior to or after completion of alcoholic fermentation.  Concurrent yeast and bacterial populations are more difficult to manage.  Yeast by-products can kill your bacteria (and vice versa), and both species compete to consume the same limited nutrient pool.  Strain factors put both populations at risk.  But then adding bacteria first, to sugary, aerated juice favors the vinegar side of the equation.  This leads to the second issue - manipulation of chemical by-products.  You know that tell-tale butter drizzled in your California chardonnay?  That flavor is secondary to malo-lactic conversion.  When and how you permit malo-lactic conversion can have an indelible impact on flavor and aroma.  You don’t want to leave this to chance.

So where exactly did I go wrong?  I think a number of factors favored team LAB (lactic acid bacteria).  These guys are stimulated by low SO2, warmer temperatures, low alcohol, and low acid/high pH, especially “the natives.”  Clearly I wasn’t aggressive enough with my sulfur dioxide at the crusher.  100mg/L SO2 is required to demolish the LAB’s.  Fast and loose, I rolled short, adding only 33.76 g SO2 to my 608.6 L of must.  (If you need to know, I derived this from hand sorting 1.005 tons of fruit yielding approximately 160 gallons/ton).  Second rookie mistake: my high tech cooling device failed miserably.  Twice a day, I lobbed a heaping scoop of dry ice into my bin to cool it down.  Next year, I’ll double scoop more aggressively, since the single-scoop got balmy.  Then I got caught in a vicious cycle.  Native LAB’s don’t like too much acid.  Generally pH<3.5 is their threshold for viability.  But I’m guessing the low sulfur allowed some natives to flourish, putting the malo-lactic machine in overdrive.  As natives converted malic to lactic acid, the pH crept higher and acid dropped, allowing more unwanted natives to sabotage my bin.  The bacteria themselves carved out a more favorable environment for their brethren.  By that point, the winery was already crawling with yeast from every other active bin in the joint, so my 5 day, pre-fermentation cold soak became a Sisyphean feat.

Like all fairy tales, this one ends with a major deus ex machina.  Happily, a big dose of commercial strain yeast set everything back on course.  After fermentation completed, I added my freeze-dried bacteria to the barrels.  The real malo-lactic conversion should be finishing up about now.  Just in time to stave the winter chill with a nice long nap in oak.