I love sparkling wine, I love Champagne and, of course, j’adore vintage Champagne.
I can never taste too often or too much and in this confessional I am a Brut Nature, Extra Brut and Blanc de Blanc aficionado.
Sometimes content finds me and not the other way around.
Let me explain.
I saw a video on YouTube that I thought at first was good but I was thrown off by an inaccuracy. Here is the inaccuracy that started me on this data quest:
“But thanks, in least, in part, to global climate change; warmer and milder years are more common in Champagne and you are seeing more and more vintages being called.”
First, I am not sure I would use the words of thanking global climate change for anything.
I should also point out what I hope is obvious is that I am not a climate change denier–just in case there is a question.
But knowing Champagne the way I do and also a proponent of science I do prefer to see the data and documentation before proclaiming something that is not accurate. Champagne has and I believe will continue to be a challenging climate–challenge is a good thing for this king of wines.
The YouTubers statement about more vintage Champagne is unfounded and an arm chair quarterbacking is not helpful but harmful. I had a hunch but I needed to back up this hunch and look at the data. I sought the data and completed a comprehensive analysis to see if there are more vintages of recent times. The overly and un-expertly view is done with a mere casualness versus a more thorough and diligent proof and building of documentation.
Here is my research – I wanted to include the oldest producers of vintage Champagne and in doing so found some of the largest producers. Here are the producers vintage wines I utilised; only one producer I used two of their vintage wines which is Clicquot.
Vintage wines represent 1-5% of all Champagne wines produced–very small and yet very telling as well.
Producers and specific wines in my study:
- Champagne Bollinger RD
- Champagne Canard-Duchêne Millésimé (total house production of 4.2 mm bottles; vintage and non-vintage)
- Champagne Dom Perignon
- Champagne Drappier
- Champagne GH Mumm Millésimé (total house production of 7.5 mm bottles; vintage and non-vintage)
- Champagne Gosset Celebris
- Champagne Lanson Millésimé (total house production of 4.2 mm bottles; vintage and non-vintage)
- Champagne Laurent-Perrier Grand Millésimé (total house production of 7.26 mm bottles; vintage and non-vintage)
- Champagne Louis Roederer Cristal
- Champagne Nicolas Feuillatte Palme d’Or (total house production of 10.8 mm bottles; vintage and non-vintage)
- Champagne Piper-Heidsieck Cuvée Rare (total house production of 4.3 mm mm bottles; vintage and non-vintage)
- Champagne Perrier-Jouët Belle Epoque
- Champagne Pol Roger Winston Churchill Cuvée
- Champagne Pommery Cuvée Louise Brut Millésimé (total house production of 4.38 mm; vintage and non-vintage)
- Champagne Salon
- Champagne Taittinger Comte de Champagne (total house production of 5.5 mm bottles; vintage and non-vintage)
- Champagne Veuve Clicquot (total house production of 19 mm bottles; vintage and non-vintage)
- Champagne Veuve Clicquot La Grand Dame (total house production of 19 mm bottles; vintage and non-vintage)
I researched for the above vintages per producer and placed them in a spreadsheet. This spreadsheet below is a “heat map” but I am careful with using this term as “heat” in this map is not to highlight actual temperature but showing where vintages land per year per producer.
This map was to show a relationship and chart what it does look like year-over-year. While I could have mapped 19th century vintage wines like Veuve Clicquot which was the first in 1810 and also Perrier-Jouët and Louis Roederer Cristal; the case of Cristal as it was produced for Tsar Alexander II and it was not available to the general public until 1945. There was not enough data to map for the 19th century and hence the biggest cluster of information is the 20th century. It might be argued that Champagne vintage wine are a phenomenon of modernity and takes a hold in the last century.
There were epic events that were also problematic in the 20th century such as WWI and WWII which Champagne was in the middle of the these wars. Notre-Dame de Reims Cathedral was bombed in WWI by Germany 1914 and received considerable damage and renovation was completed in 1939. In WWII, the surrender of Germany took place 7 May 1945 at this former school which is now the Musée de la Reddition in Reims (Museum of the Surrender); when I stayed in Reims in 2016 my hotel was across the street from this museum.
But even during both world wars Champagne was still being produced and few vintages were produced. Even with difficult conditions people still endured to produce wines in Champagne.
Back to the heat map of Champagne vintages. I thought this was a good starting point to see what the vintages against each producer and year might look like. This map charts 1921 when there are three producers with vintages in that year. This data collection has 90 years of data (I did not include 2010s because not all vintages are declared or known yet). But I think this set of 90 years worth of data was adequate to look at behaviours of Champagne vintages. There are two notable colours in this map: orange showing vintages for all producers in one year (i.e. meaning all producers had a vintage for particular year) and blue denotes no declared vintages for all producers in a given year.
There are only three years where all producers had a vintage on the same years – those years are 1921, 1928 and 1934–this unto itself is not remarkable as there are only three producers those particular years. But more interesting is that no other year after 1934 had an all vintage year by all producers. If there were more vintages year-over-year as is supposed by arm chair quarterbacks all producer vintage years would happen again and if not commonly.
The blue represents years where no vintages took place in a given year – those years are 1922, 1924, 1927, 1930, 1931, 1933, 1935, 1936, 1938, 1939, 1940, 1941, 1944, 1950, 1954, 1956-1958, 1963, and the last was observed in 1968. The 1930 and 1940s isn’t too surprising given few producers and this is a WWII period and hence little demand. The 1950s ushers in 5 vintages from a total of 9 producers with a vintage wine. But looking at 1968 and even earlier it is hard to recognize a pattern and hence a numeric analysis is needed.
Also, keep in mind there could be many more producers with a vintage especially in the last half of the 20th century–they were not included simply because it is especially difficult to account for every single vintage by all producers–such data is not readily available or findable. Perhaps the only body that could do this is the CIVC (Comité Interprofessionel de Champagne). I do think this data set is relevant and important as many of these large producer have well known wines and well known vintage wines.
Another way to view this would be to look at total bottles of vintage wine produced by all producers; the difficulty there is to understand that quantity to weight the average of all vintages if possible–again this data would be very difficult if not nearly impossible to assemble.
Some top line data points:
This study shows 411 vintage wines for the 20th Century of these producers and if I add 19th Century then there are a total of 423 vintage wines. Vintage range per producer in this study are from 7 vintages to 64 vintages (only one producer can say this and that is Veuve Clicquot). If all producers are averaged for total vintages produced that number is 22.83 or 23 vintages in total. When you add Clicquot which is important to do–the total span of vintage possible years is 200 if measure to 2010.
The span of vintage production ranges from 30-200 years from first to last vintages produced to the year of 2010 – the average if you use Clicquot is 59.5 years and if you remove Clicquot the average is 51.2 years.
The average time between vintages for 20th century production is 2.56 years meaning that on average there are four vintages per decade per producer.
If there were more vintages being produced this would be most pronounced especially since say 1980, 1990, 2000 decades (or even earlier) if global climate change was inducing more vintages. The gold curve represents total vintages possible per each given year (more precisely if each producer had a vintage every single year it would be the gold curve).
The blue curve show the total vintages per each given year. This curve would have fewer valleys and it would be reaching to gold curve if not touching this line if Champagne was experiencing increased growing degree day temperatures. In fact, 2006 to 2010 shows a decline in the number of vintages.
I didn’t include the earlier period with three producers each having the same year vintages: 1921, 1928 and 1934–as this behaviour didn’t repeat itself. So I looked for years where there were highest percentage of vintages. The above line graph captures these vintages and they are as follows:
1985 and 1988 is the only decade with two of the highest percentages in this study of 20th century vintages. There is a decade gap between 1966 and another gap in decade for the 2004 vintage. The 2004 vintage is only 1 percent higher than the 1985 and 1988 vintages. These high percentages would be at this level especially in the 2000 decade but it is 16 years if we include our current year of 2020. Wouldn’t we see more vintages in the late 80 and 90 percentiles as evidence of global warming–if those conditions were increased growing degree days in Champagne?
This intriguing line graph starting from 1900 to 2010 depicts average vintages by decade and producer. This line graph highlights a large amount of historical data and each line is a different producer. While this seems erratic–it is and it would be disappointing if it weren’t–Champagne has an irregular climate to produce routine vintage wines.
The orange line is integral as it is an average of all producers. The dotted purple line is the trend line which is showing an upward tendency i.e. increased vintages from 1900 to 2010–while the answer is yes from a longer historical period of 110 years it might be the answer one might use to show more vintages of recent times.
While the above graph is important–I thought that a more focused view would be an interesting and a valuable view. This graph below depicts 1960 to 2010 – the longer view did have a particular and predictable trend line because it would show an upward trend–it would have to for very logical reason–more producers entering with vintage wines. And hence a trend line loses value in this example because it is capturing entrants in vintage producers as well as increased vintages.
Hence the 1960 to 2010 vintages has a normalized trendline because this cut of data shows a majority of vintage producers. The trendline has flatlined if not slightly tilting downward. This graph below does not show an increase number of vintages which I believe is accurate. Champagne is as unpredictable in this study period as one might expect and it does fit the reputation of Champagne.
When the next decade of vintages becomes available and when added to this current data set will it show if there is an increase, decrease or equivalency–that will be interesting to see.
The CIVC does not declare a vintage–a producer does–the CIVC does have requirements for a vintage wine such a minimum of three years aging in bottle and a house that is a mix house of non-vintage and vintage wines cannot use all of their wines produced in one year solely for vintage labels–wine must be saved as reserve wines.
I suspect that in the entire history of wine making in Champagne there has never been a decade where a vintage wine could have been produced in every single year of any decade. What keeps a Chef de Cave from declaring every year a vintage wine? Simply reputation–no Chef de Cave would risk their reputation along with their house. Reviews of a producer where there are constant year-over-year vintages would show in those scores.
Champagne as I suspect even with global climate change will not produce more vintages–in fact the opposite could also become a reality: potentially fewer vintages.
Why potentially fewer?
If there is more rain, wind, colder weather, less sun, more frost, late budding, reduced growing period, etc. i.e. challenged growing degree days. Global climate change is not always or only about increased temperature but much more than that.
Champagne is complex (both wine and region) for several reasons–it is at the frontier of a northerly latitude: 49°5, a predominate oceanic climate with continental accents. But increased global temperatures will not be a 1-for-1 ratio of average global heating units will not evenly translate to Champagne. I do think and I have believed that Champagne is at an intersection of challenge and beauty. I speak glowingly about one of my favourite wine regions on the planet–I love the story and I love what difficulty can produce: greatness, elegance and finesse lyrically served by the glass.
Here is my podcast on this topic on Spotify.
James the Wine Guy
Special thanks for the CIVC – Comité Interprofessionnel du vin de Champagne, Champagne Louis Roederer, Champagne Veuve Clicquot, Champagne Lanson and Champagne Drappier.
Demystifying Wine…One Bottle at a Time from all wine regions around the world.
© 2020 James Melendez / James the Wine Guy— All Rights Reserved – for my original Content, drawings, art work, graphs, photographs, logo, brand name, rating, rating, taxonomy, graphic and award and designs of James the Wine Guy. James the Wine Guy is also on Facebook, Twitter and most major social medias.
Follow, subscribe, like, browse: