Harvest 2017 Recap: Fast Start and Finish, Solid Yields, and Plenty of Concentration

Last week, with the threat of our first rainstorm on the horizon -- key word here is threat, as it didn't pan out -- we brought in the last lingering blocks of Mourvedre, Roussanne, and Counoise.  And then, all of a sudden, we were done.  Sure, there are a few lots we're still waiting for, as they sit concentrating on straw in our greenhouse. But the picking, at least, is through. No wonder our cellar crew was ready to celebrate:

The harvest came in two large waves, with a significantly less busy stretch in between.  The first surge was kicked off by a record heat wave that sent temperatures soaring over 102°F nine days in a row in late August and early September, and included the majority of our purchased grapes for the Patelin de Tablas wines.  The slow interlude corresponded with nearly three weeks of temperatures more than 5°F cooler than normal. When, by early October, things heated up again (modestly this time) most of the late-ripening grapes on our estate were nearly ready, and we had a pretty continuous run between October 2nd and October 19th, picking 16 of 19 days in that stretch. You can see the double-peaked workflow in the chart below. In the chart, blue is purchased fruit for the Patelin program, and orange estate grown fruit:

Yields were the best we've seen since the twin vintages of 2005 and 2006, up about 20% from 2016. On the one hand, this should be unsurprising, given that we got good rainfall last winter after five years of drought. But you still never know, and midway through the summer, we were all thinking that harvest was going to be at average or slightly above average quantities.  To have it among our most productive vintages ever was somewhat of a surprise. The complete picture:

Grape	2017 Yields (tons)	2016 Yields (tons)	% Change vs. 2016
Viognier	18.9	14.2	+33.1%
Marsanne	13.8	4.5	+206.7%
Grenache Blanc	46.4	30.6	+51.6%
Picpoul Blanc	9.7	7.7	+26.0%
Vermentino	22.2	19.0	+16.8%
Roussanne	41.7*	47.0	-11.3%
Total Whites	152.7	123	+24.1%
Grenache	73.1	58.8	+24.3%
Syrah	41.5	36.8	+12.8%
Mourvedre	72.9	62.7	+16.3%
Tannat	20.5	12.3	+66.7%
Counoise	18.8	18.0	+4.4%
Total Reds	226.8	188.6	+20.3%
Total	379.5	311.6	+21.8%

Overall yields ended up at 3.62 tons per acre, about 20% above our ten-year average.  Other years in which we've seen yields around 3.5 tons per acre have included 2005, 2006, 2010, and 2012, which includes some of our favorite vintages. Like 2005 (the vintage that this reminds me the most of) this year followed a multi-year drought, and the obvious health of the vines suggested that they were able to ripen the relatively healthy yields with concentration and character.

*If you're wondering why Roussanne has an asterisk, there are two reasons. First is that there's still a little Roussanne concentrating in our greenhouses, to be added to the total. It won't amount to much (a ton, more or less) but it's there. The second reason is that it's worth noting that at least some of the decline comes because we pulled out an acre or so of Roussanne this winter. We also pulled out a couple of acres of Syrah, but the Syrah total is augmented by the fact that the year before, we grafted over about two acres of Roussanne to Syrah.  So, Roussanne has seen a bit of a double-whammy in recent years. We'll be planting some more Roussanne soon.

Another way that you can get a quick assessment of concentration is to look at average sugars. Since 2007, the average degrees Brix and pH:

Year	Avg. Sugars	Avg. pH
2007	24.42	3.67
2008	23.87	3.64
2009	23.42	3.69
2010	22.68	3.51
2011	22.39	3.50
2012	22.83	3.65
2013	22.90	3.63
2014	23.18	3.59
2015	22.60	3.59
2016	22.04	3.71
2017	22.87	3.74

You'll note that 2017's sugars saw a rebound after the lower levels in 2015 and 2016. This is a sign of the health of the vines. The higher pH levels seem to have been a result of the stress that the early-harvest heat wave produced, as well as a reflection of the overall warm summer.  In the 2017 growing season, only May was cooler than average, with April, June, and July particularly warm.  Overall, 2017 was our second-warmest year ever, just a hair cooler than 2014 and about 10% warmer (measured in degree days) than our 20-year average. The chart below summarizes (October's information is for the first 19 days, as we picked our last significant block on October 19th):

We picked even more lots this year (124) than last, and we ran out of space on our harvest chalkboard.  Note the number of times that you see Roman numerals after a pick, particularly toward the end of harvest: those are the blocks that we picked multiple times:  

The duration of harvest -- 54 days -- was exactly at our average this millennium. Both the beginning (August 30th) and the end (October 23rd) were a little earlier than our average since 2000, but not by much: about 4 days.  That said, we started later than we have since 2012, and finished later than 2013, 2014, and 2016. The fact that 2017 saw a later budbreak than recent years pushed everything into a more normal time frame than we've seen recently, but it was still a warm year.

In character, it's early to tell what things will be like, but so far, we're excited. I asked Winemaker Neil Collins to sum up the vintage based on his tastings, and his response was, "full, and great structure, and lovely acidity... just what we wanted it to be."  Given that Neil, like most winemakers, tend to focus on the shortcomings they see in the immediate aftermath of harvest, this is a pretty resounding endorsement.  We'll know more, of course, in coming weeks.

The last project for us for harvest 2017 is to make our first Vin de Paille "Sacrérouge" since 2014.  This traditional dessert wine-making technique involves drying newly-harvested Mourvedre clusters on straw, and only pressing and beginning fermentation when they have reached our desired level of concentration -- typically around 35° Brix -- after 2-3 weeks.  [For more details on how and why it's done, see our blog Vin de Paille: A Dessert Wine Making Technique for the Obsessed from a few years back]. The Mourvedre we'll be using for this project is currently sitting on the straw, and we expect to bring it in and start fermentation later this week. 

Now that the fruit is in, it's welcome to start raining any time.  Meanwhile, we're enjoying the autumnal views of the vineyard without having to worry that the cooler nights and the likelihood of future precipitation -- our first chance, it appears, may come as early as the first few days of November -- might negatively impact our harvest.

Harvest at the Three-Quarter Pole: A Return to a More Normal Time Frame, with Solid Yields

Late last week, we welcomed our first major picks of Roussanne and Mourvedre into the cellar.

And with that, the home stretch of harvest officially began. There will be a lot of harvest chalkboards that look essentially like this one over the next couple of weeks:

Where we are, one week into October, is remarkably similar to where we'd expect to be, if we were predicting at the beginning of the year.  We're done with early grapes like Viognier, Vermentino, Syrah, Marsanne, and Grenache Blanc.  We're mostly done with what we consider mid-harvest grapes like Grenache and Tannat.  And we're just getting into our late grapes, Mourvedre, Roussanne, and Counoise.  Given that we're comparatively heavily planted in these late grapes, we still have more fruit out than many of our neighbors.  Still, we expect to be harvesting pretty steadily for the next two weeks, and to be done before the end of the month.  If this seems late, it's likely a matter of perspective, because most of our recent years have been early.  While 2013, 2014, and 2016 were all done by mid-October, our average finish date of harvest this millennium has been October 29th.

With the first complete blocks harvested, we've been able to get the animals back into the vineyard.  Right now, they're in the head-trained vines on our Scruffy Hill block, visible from Vineyard Drive if you're coming in from the south:

Although we're where we'd expect to be in the harvest sequence, it hasn't always been smooth getting here.  Harvest began with a significant heat wave that sent temperatures soaring over 102°F nine days in a row.  We then got nearly three weeks of temperatures more than 5°F cooler than normal. In the last two weeks, temperatures have been more or less normal for the season, without any noteworthy heat waves, and with only one day significantly cooler than normal, a bizarrely chilly October 3rd where the sun didn't break through the fog until noon and the day topped out at 64°F:

For the month of September, we had 11 days warmer than seasonal averages, and 19 days cooler than average.  Even with the heat wave that began the month, our average high was 86.3°F, two degrees cooler than average. These cooler days allowed the vines to recover from the stress of their early-season heat wave, and allowed the cellar to free up tanks and get ready for the next push.  A graph of the harvest by week shows the ebb and flow. Normally, you'd expect a sort of bell curve, with thin tails at the beginning and end and the busiest weeks in the middle.  Not this year:

In terms of yields, with a significant number of grapes done, things are coming into focus.  It looks like yields are up from 2016, and a bit above average for the first time since 2012.  The varieties we've finished harvesting are up an average of 32.9%, with the most noteworthy recovery from Marsanne, whose yields had been so depressed by the five years of drought that we were getting less than one ton per acre last year:

Grape	2017 Yields (tons)	2016 Yields (tons)	% Change vs. 2016
Viognier	18.9	14.2	+33.1%
Marsanne	13.8	4.5	+206.7%
Grenache Blanc	46.4	30.6	+51.6%
Picpoul Blanc	?	7.7	?
Vermentino	22.2	19.0	+16.8%
Roussanne	?	47.0	?
Grenache	73.1	58.8	+24.3%
Syrah	41.5	36.8	+12.8%
Mourvedre	?	62.7	?
Tannat	18.3	12.3	+48.8%
Counoise	?	18.0	?
Total so Far	234.2	176.2	+32.9%

Even with the higher yields, sugars are up a bit this year, which is a sign of the health of the vines.  Thank you, rainy winter!  The growing season, the yields, and the character and numbers of the grapes at harvest remind us most, so far at least, of 2005: also the first wet year after a string of dry years, with a long growing season and a relatively cool harvest period.  We aren't likely to go as late as we did that year -- November 7th -- but if we get a similarly robust vintage, we'll be happy. 

Meanwhile, we'll enjoy the last couple of weeks of grapes on the vines. By the end of the month, we'll have to wait another year for views like this:

Why Paso Robles is So Well Suited to Late-Ripening Grapes

This morning, when I got back to the winery after a week on the road, my first order of business was to check in on how harvest was going. I was happy to learn that things picked up a bit last week. After more than two weeks of chilly fall weather, it had warmed back up, with eight days of perfect ripening weather: daytime highs between 83°F and 93°F, and lows between 41°F and 51°F.

And still, when I asked Chelsea how she was feeling, she responded, "this is definitely the first October 1st I can remember where we haven't been stressing about tank space."  Although harvest picked up from the glacial pace it was in mid-September, we are still waiting on most of our Marsanne, Roussanne, Mourvedre and Counoise.  Why? Blame the cold nights. Here's Neil, this morning, next to our first pick of Mourvedre. It was 52°F at 8:30am:

We're used to this here, but most of the Mediterranean world is finishing up harvest about now. Beaucastel's Facebook page (for example) shows that they brought in their last fruit on September 29th:

It's not like this year is an outlier for us, either.  Over the last 15 years, we've averaged a last pick off the estate on October 29th, and our earliest-ever finish was October 7th in 2013.  Six times in those 15 years we were still picking in November. 

To explain why grapes take so long to ripen in Paso Robles, I'll have to detour briefly into some basic plant physiology. Bear with me here, or just skip to the end of the bullet points if you'd like the conclusions without the chemistry. There are a number of different processes which limit a grapevine's ability to photosynthesize at low temperatures. These include:

• The tendency of plants to close their stomata (pores in the leaves) in response to cold, limiting respiration and the uptake of CO2
• Carboxylation (sorry for the long, technical term) is the first stage of photosynthesis, whereby CO2 molecules are turned into an acid known as 3-PGA. Carboxylation efficiency declines as temperature declines
• The electron transport capacity of plants is reduced at low temperatures
• An enzyme known as Rubisco, essential to the first step of carbon fixation in photosynthesis, is inefficient at low temperatures

So, in essence, at cold temperatures, plants take in less CO2 and are less efficient in turning the CO2 that they do take in into the starches that fuel both plant growth and fruit ripening.  Grapevine ripening proceeds most efficiently between 30°C and 35°C (86°F and 95°F).  It drops dramatically below 25°C (77°F), and reaches zero at 10°C (50°F).  A summary graph from a technical paper published in Plant, Cell, and the Environment shows the combined effects pretty clearly:

For context, take a look at the temperature curve for the most recent 24 hour period:

You can see that while it did get warm, topping out around 30°C (86°F) yesterday afternoon, it only lasted until sunset just after 6pm.  By 8pm it was already down to 20°C (68°F). It bottomed out at 6.4°C (43.6°F) at 6am and wouldn't rise back up above 20°C until noon today.  So, over the last 24 hours, our vineyard spent 5 daylight hours over the 25°C temperature at which photosynthesis happens efficiently (2pm-6pm yesterday). Five other daylight hours (9am-1pm today) saw temperatures at levels where some photosynthesis can happen. Two daylight hours (7am-8am today) saw no photosynthesis at all because it was too cold.  And for 12 hours the sun was below the horizon. 

We are far from the only, or even the most extreme, location in Paso Robles.  The temperature grid from the Paso Robles Wine Country Alliance shows other areas that dropped near freezing last night.  Most show diurnal temperature swings of 40°-50°F. 

So, what does all this mean? That once you get into the end of the growing season here in Paso Robles, it's hard for grapevines to do too much photosynthesizing. That's a benefit, because you can get the last little bits of ripening on your late-ripening varieties slowly, so they continue to build complexity without accumulating too much sugar.  In general, the longer your grapes can stay on the vine before they get to the ripeness levels you want, the more complexity your wine has.  That's why a generally accepted bit of wine wisdom says that the best examples of different grape varieties can be found at the northern limit of their ripening range. So, the best Sauvignon Blancs tend to come from the Loire, and not Bordeaux. The best Pinot Noir tends to come from Burgundy, and not the Languedoc.  And the best California Chardonnay tends to come from cool coastal pockets where the fog slips in from the Pacific, not from the Central Valley.

Of course, at some point, you do need to get things ripe.  Grapes that don't make it to good ripeness produce wines that are green and bitter: no one's idea of a pleasurable drink. But here too Paso Robles has an advantage: that we don't tend to get our first serious rain until mid-November.  If we need to wait, we wait.

Hopefully, this particular waiting game is over for a while. But if it's not, I'm still confident we'll be OK. Thanks, Paso Robles.