subscribe: Posts | Comments      Facebook      Email Steve

Talkin’ AVA boundary blues


Once again the subject of new AVAs, or American Viticultural Areas, arose in conversation this week, relative to a story I’m working on. It concerns an area that some people, including me, as well many if not most who actually grow grapes and have wineries in the region, believe should have its own unique appellation. I won’t mention it, because you’ll read about it in an upcoming issue of Wine Enthusiast.

The problem, as one person expressed it, comes down to this: “You either have to cut cut some people out, or include everybody, to the point where it’s almost meaningless.”

He’s talking about boundaries, of course. They have to run somewhere, have to be based on something (climatic, dirt-wise, watershed, history). They can’t just be arbitrary because that would pollute the very concept of an appellation. Trouble is, depending on where you draw the boundaries, somebody who wants in always gets left out. (Never mind that some people who shouldn’t be in automatically get included. That’s a different story.) Whoever gets left out becomes grouchy and, potentially, litigious. And who wants a grouchy, litigious neighbor? (As a condo dweller, I can say with certainty that I don’t!)

I thought this topic was worth some observations on some current California AVAs. I won’t mention all 9,476,746 of them because then this post would be War and Peace. But here’s a start.

All the valleys need to be redefined to bring down their elevation lines. For example, in Sonoma Valley,  the AVA line goes up to 2,530 feet! That’s a half-mile. If 2,530 feet is a valley, then I’m a [fill in ridiculous metaphor].

(All my statistical conclusions about AVAs in this post come from the Wine Institute’s link to American Viticultural Areas.)

Alexander Valley’s highest permitted elevation is 857 feet. Again, that is not a valley–conditions that high are in stark contract to the valley floor. These higher stretches of Alexander Valley (even well above 857 feet, where I guess they’d qualify only for Sonoma County) desperately need their own AVA. The late Jess Jackson tried, and failed, to get Alexander Mountain an AVA. I don’t care what the name is, but somebody should continue that struggle.

Anderson Valley goes up to 2015 feet feet, a situation relieved, in some part, by the Mendocino Ridge appellation. Still, it’s weird.

Many of California’s AVAs really don’t matter because few wines are produced from them and much of that is common. The bigger AVAs, like Napa Valley, are pretty meaningless, especially in this age of the negociant, where tons of wine with Napa Valley on the label are standard stuff. Napa’s mountain AVAs make sense to me. As for the communes, well, they’re weathervanes, in the sense that an Oakville Cab will have some particular characteristics, a Rutherford Cab will have its own traits, and so on. But these generalizations often break down under close inspection. I think they need to split Oakville in half: east and west. That would be a beginning. Rutherford’s awfully sprawling. It takes in everything from Flora Springs on the west to Staglin in the middle to Quintessa with its little hill there to Hall up above the Silverado Trail in the Vacas. Crazy. And don’t even get me started about Russian River Valley. I’ve been calling for years for it to be subdivided into at least three AVAs, maybe as many as six. Nothing ever seems to happen.

Edna Valley–now there’s an AVA that makes sense. You can tell an Edna Valley wine a mile away, whether it’s red or white. Green Valley’s pretty dependable, too. Santa Rita Hills, Dry Creek Valley–you usually know what to expect from them, and you get it. Those are AVAs at their best.

I do understand the argument about grouchy neighbors being left out. That’s too bad–it’s why so many of our AVAs are so porous. They let in vineyards that shouldn’t have been there, out of consideration (or fear) of somebody. Still, we need more appellations in California, not fewer, but they should be rational and small, not gigantic and meaningless. San Francisco Bay? A lovely place, my home, but not an appellation, in any sense except that somebody had enough money to buy it.

What are your nominations for the smartest and dumbest AVAs in California?

  1. AVA’s are supposed to be a regions, large or small, that have similar growing conditions. It is supposed to make people think that it’s an area that has, for the most part, a similar terroir throughout. The purpose being to further define the growing conditions of the grapes. Like anything else they are made famous by the wineries in them. Houses are the same way. “Did you know “John Steinbeck lived in this house”or “George Washington slept here”
    With so many wineries growing the same varieties under such varying growing conditions and many of them not only well made but quite good I don’t see any but the most esoteric of wine enthusiasts really paying attention the the particular AVA. I here people say Napa or Sonoma or Paso a lot but not further defined from there. What people should know is that Santa Clara Valley is the best climate in the world to grow grapes.

  2. Dumbest — Sonoma Coast. It’s worse than RRV, I think. Forces all kinds of winemakers to say their wines are from the “true” Sonoma Coast. I’ve heard all the stories about the Petaluma Gap, but no way all those inland vineyards around 101 should be in the same AVA as those up on the North Coast.

  3. Paso Robles – I wouldn’t call it the dumbest AVA, but perhaps the most in need of subdivision. Let’s at least split it down the middle with US 101. This would separate the more arid Salinas valley on the east from the wooded foothills of the coastal range on the west. Of course there is room for a lot more segmentation beginning with the “Templeton gap” area where CA 46 heads west up to the York Mountain AVA (a story in itself). From what I recall, there has been discussion in the past of dissecting the Paso Robles AVA into as many as 14 sub appellations. I wonder if anyone has an update on this concept?

  4. While the geographic heart of most AVAs are usually spot on in capturing the historical and cultural import and to a lesser degree shared organoleptic markers (grape tannin profile, weight, aroma and flavor profile) exhibited in grapes/wines from the AVA, it is at the borderlines of the AVAs that they fall short, and tend to lose their value as a good indicator of certain flavors/tastes/aromas. And it is at the geographic borderlines of many AVAs where all bets are off. Any romantic notions I may have harbored re: consistency in the accuracy and scientific-cultural basis for AVA delineation were vanquished after a conversation I had in 2000 with David Clark (winemaker and co-owner of the old Deer Park Winery, which was located about mid-slope on Howell Mountain). Clark said (in words or substance) that shortly before the Howell Mountain AVA was established, “the powers that be” told him that he could choose to have the boundary for the lower portion of the forthcoming Howell Mountain AVA drawn so as to include his estate vineyard within the new AVA, or not, his choice. So much for science. Same with the expansion of the border of the Russian River Valley to include Gallo’s Two Rock Vineyard. Anyone who’s done nightpicks in the heart of the Russian River Valley (River Road area between Olivet Road and Trenton-Eastside Rds.) and also down near Stony Point Road/Cotati where Two Rock Vineyard is will tell you that it’s an order of magnitude colder and breezier at night that far south, especially in late September. I firmly believe that vinifera grapes grown in the same or similar soil types in the same neighborhood often show similar qualities in terms of tannins, structure, weight, and sometimes flavor and aroma, and therefore a smaller, more focused AVA or sub-AVA can have real meaning for consumers and to our cultural legacy as a wine producing nation (Howell Mountain and Green Valley of the RRV standout in this regard) by serving as a “quasi-source indicator” or geographic trademark. But when we expand the AVA borders to go beyond the core neighborhoods or ridgelines that are at the heart of that AVA, we tarnish the meaning and quasi-source-indicator function of the AVA, and diminish the value of the AVA to all parties involved. (And since you can now amortize the value of an AVA as a separate Section 197 amortizable asset acquired along with a vineyard property, the diminishment in the value of the AVA over time presents all kinds of issues beyond the loss of meaning to consumers.)

    Best regards,


  5. Greg Jones says:

    I have done much research on AVAs, including digitizing all of the boundaries in the western US (>130) and analyzing area, elevation, climate, geology, soil, etc. Most of this work appears in the AJEV paper referenced below. Also note that the AVA areas and elevations in this publication are more exact than the Wine Institute’s tables, but are only as were approved by 2008 (so there are some new ones and some changes to existing ones).

    Jones, G.V, Duff, A.A., Hall, A., and J. Myers (2010). Spatial analysis of climate in winegrape growing regions in the western United States. American Journal of Enology and Viticulture, 61:313-326.

    From doing years of work on all of these AVAs I can say:

    1) The TTB process for having these submitted and approved is not very good, verging on flawed. Petitioners basically draw the lines on old USGS quad maps and give metes and bounds descriptions of the lines. First, this is so old school as all of these boundaries should be submitted as electronic, geographically projected, vector file lines that can be mapped accurately. I have argued this with the TTB to no avail. Second, this process results in numerous errors that come from old, out of date USGS maps and wrong translations from the map to the metes and bounds descriptions. I have found errors in many approved AVAs.
    2) The other issue is that these boundaries are usually petitioned to follow some cultural geometry (streets, railroads, township/range lines, etc) or physical entity (river, creek, ridgeline, moutain peak) that have no basis for being a true wine region boundary. These are mostly done based on convenience and somewhat done based on perception of how the petitioner sees the area … or in being inclusive or exclusive relative to growers/producers.

    In my work on examining suitability factors in AVAs I can tell you that there are numerous areas within most AVAs that will never be planted due mostly to elevation. The boundaries just encompass too much area or the wrong areas. Given our current digital data availability, one could completely redo all of the AVA boundaries, based on elevation and climate, and produce something quite accurate. But the new lines would not be pretty.

    Just examining the climate in the AVAs in the western US, one can find up to 4-5 different climate types capable of ripening many varieties and producing a wide range of wine styles, in one AVA … does this make them uniquely tied to the terroir of that region or are they different?

    As for which current AVAs make sense or not, I would say that it is in the eye of the person trying to sell the wine. For example, some uber AVAs make sense complete sense, but only in the marketing of the wine (e.g., Columbia Valley, Willamette Valley, etc). While some very small areas make sense in terms of being unique geologically, climatologically, etc they only serve a few growers/producers (as a matter of fact, I could easily argue the scientific merits of numerous new small AVAs all over the west).

    Hope this adds something to this discussion …

  6. Steve wrote: “Alexander Valley’s highest permitted elevation is 857 feet. Again, that is not a valley–conditions that high are in stark contract to the valley floor. These higher stretches of Alexander Valley (even well above 857 feet, where I guess they’d qualify only for Sonoma County) desperately need their own AVA”.
    True. Despite the mere descriptive nature of American AVAs, in contrast to the prescriptive nature of the French AOCs, it should also be noted that, even though two weather stations from the North Coast AVA (e.g., Mendocino Ridge, Sonoma, Napa, etc.) and/or the Santa Cruz Mts., can exhibit the same (Winkler) Heat Summation Index number, there is a lower diurnal fluctuation (i.e., lower highs and higher lows) at higher elevations when compared to valley floor sites.
    In addition, the growing season’s accumulated solar radiation (in those areas) is lower at higher altitudes; however, it generally reaches higher average levels than the valley floor during the summer months. This means that the season gets shorter as the altitude increases. Thus, the sole impact of altitude on grapes’ flavor profile is unequivocal.
    Although the Winkler/UC-Davis Heat Summation Index cannot address some important features present in the data, various other “heat” indicators can do the job quite effectively; and a plain comparison of climate data from Glen Ellen (, Eldridge, Larkfield-Wikiup, Boyes Hot Springs, Healdsburg, Cloverdale, Ben Lomond1 (alt. 450 ft.), etc., with data from Hawkeye-Mayacamas ( and Ben Lomond2 (alt. 2,633 ft.), seems enough to validate the argument above.

  7. Peter, higher elevations are also generally above the fog.

  8. Steve,
    Absolutely. And since (sea) fog is fundamentally a late-spring/early-summer phenomenon, it largely explains the difference in the seasonal distribution of sunshine and heat between mountain (above the fog-line) and valley-floor sites in the North Coast; and why the shape of the distribution gets peaked (leptokurtic) as the altitude increases, and skewed to the right (i.e., towards the end of the season) in the lower coastal areas.

  9. Brian Cheeseborough says:

    The Sierra Foothills is a massive geographical territory and all of the the subdivision are drawn on county lines (El Dorado and Fairplay being exceptions) Most of the wines use political boundries…Amador, Calaveras, Mariposa, I believe there is 6 or seven counties in all. As a winemaker, being able to use a political designation for “AVA” is nice you can use more,up to 15% of another vintage(s) to complex the blending.

    I also agree Paso Robles should be broken into more than one AVA, probably 5-7. Their is a significant difference between Adelaida Road on the westside & Hog Canyon in San Miguel

  10. Regarding Paso Robles, splitting it down the middle at the 101 would probably cause more harm than good. Westside Paso Robles is as varied as the rest of the appellation, having both the warmest temperatures as well as the coolest. Not to mention that the Templeton Gap doesn’t stop when it hits the freeway, it continues on down into the Creston area. Paso is a huge AVA and it would be nice to have sub-divisions, but it needs to be more thought out than “Eastside vs. Westside.”

Leave a Reply


Recent Comments

Recent Posts