2003 Bordeaux Vintage Report - The Insider's Guide
Every year, Bill Blatch of Vintex SA, a leading Bordeaux Negociant, sends us his detailed vintage report that he has painstakingly compiled throughout the growing season. By accumulating endless snippets of information from growers and owners on site in the vineyards and adding these to his recorded weather condition, Bill produces an intriguing 'insider's guide' to the potential quality of the wines. This, together with his diary of the vintage follows:
Preliminary Bordeaux 2003 vintage report November 16th - December 23rd 2003
France has had only eight vintages this century with such an extended period of extreme heat, and never ever before has it culminated in such an outrageously fierce August burn. It should have been enough to dry out the whole French vineyard, grapes and all. Yet Bordeaux seems to have fared better than elsewhere, first by having had a good cold and wet Winter, and then by gradually leading up to that extreme August by a hot May, and an even hotter June and July. In addition, the peaks of heat were less violent and less absolutely dry than elsewhere in France, with the result that only some of the vines actually shut down and the grapes could keep on evolving, in spite of the extreme conditions. The other capital difference from the rest of France was that Bordeaux, right in the midst of all this heat, actually got some good rainfall in July
The harvest started exceptionally early, in the extreme heat of 12th August, the earliest since 1893. A week or two later, once the rest of France had just about finished, after the re-invigorating late Summer showers and cooler nights, those whose grapes hadn't actually totally dried out in August or who had not been panicked into premature picking, could wait for true phenolic ripeness and harvest at leisurely pace, all through September.
The result is two kinds of wine, the first burnt into shape by the extreme conditions of August, the second evolved in a more relaxed way thereafter. Both are as unusual as the year's climate. Nobody had any experience of handling such an extreme vintage, and even less of assessing its results. The sugar-levels seemed so high and the acidities so low, even before the grapes were fully ripe, that many lesser estates took fright and harvested right there and then before any more yield would get evaporated away. Such wines bear the '03 hallmark of high alcohol yet are light in colour and flavour. But those with more heat-resistant soils, older vines on good rootstock, and a well-kept vineyard were able to hold off, could enjoy watching the acidities at least partially return in the determining fine but cooler September conditions, and enjoy ripeness as well as alcohol, which of course was high, but nowhere near as outrageous as the "aoûtiens" had jealously warned.
At first, these later harvesters were shy to say anything about the quality of the vintage, having just lived through all the hot-vintage nightmares of cooling difficulties during vinification, of stopped fermentations and the ever-present risk of volatile acidity, and seeing a result that, because of its exaggerated alcohol and very low acidity, was not at all their, or their oenologists', idea of what Bordeaux ought to be.
Today, often under the influence of admiring outsiders, they are gradually warming to it, and beginning to admit at least certain comparisons to other extreme vintages which ended up becoming famous. Of course, as everywhere in France, the yields are horribly low, and the moment when these wines will be put on the market is only a few months away, so they are not averse to such rumours. But most remain unconvinced that this is a truly great vintage. It is still early days and the wines are not yet really tasteable, but the following account of the history of the vintage although pointing to many rather average wines, also explains some of the really outstanding successes, surely to become great wines.
Winter 2002-2003
Although in Bordeaux we had the impression that it was a long, cold Winter, it was in fact a perfectly normal one, mild and normally wet up to the New Year, then very cold and still normally wet for January and early February.
Rain fell very regularly and slowly all through the Winter, allowing the water to penetrate the soil well and building up reserves that were to be vital later on. This is the main difference to the previous similarly hot-Summer vintages '89 and '90 which were preceded by an arid Winter. As soon as the last 2002 grape was picked on October 9th, after one of the driest harvests we could remember, it started to rain, with October about on the average at 74mm, November and December much wetter with 115mm and 91mm, and, in spite of prolonged cold snaps, January registered 93mm, again right on the average. It was only February that was down, at 57mm, and most of that fell early in the month (This was when the upper Dordogne had all that disastrous flooding). From early February until mid July, there was to be little more serious rain, but a lot had already fallen and the water-tables were normal going into the season.
The mild temperatures of the last 3 months of 2002 were not a problem; everyone much prefers to have it mild early and cold later rather than the other way round, with all the attendant frost risks of the latter. With the sap falling only slowly, it was difficult to find good days for pruning outside of early December, and this provided a good excuse for growers to put it off until much later - the real reason being that yields had been short in 2002, and there was no point in taking undue risks with the 2003. (later pruning reduces risk of frost damage)
After the violent storm of January 2nd, a lesser preview of what was to come on June 24th, there were two very cold weeks with 10 nights of frost and one day (January 10th) when it was -5° all day, followed by 3 more cold snaps: January 27th-31st (when the UK's airports closed because of snow), February 6th-10th and February 12th-19th, a total of 22 days of frost. This is a lot for Bordeaux, and the mean temperature for Jan-Feb ended up 1° colder than the average.
It meant that the sap could really go right down, and assured that parasites got killed off. If we have to have global warming (and for the Globe, 2003 was +0.45°C warmer than the 40 year average, making it the 3rd warmest year since records began), it is better for us that it comes in such long cycles, rather than in short bursts.
Spring, budding and early growth
Then suddenly, the sap started to rise when the warm Southerly winds of February 20th-28th provoked a few frighteningly early swellings. Some heavy showers March 1st-5th were followed by temperatures in the 20s March 9th-12th, these swellings became widespread and most bud-bursts happened quickly during the warm dry days around March 20th, a full week or 10 days ahead of schedule. The shoots sprouted almost immediately, influenced by some light showers March 27th-28th and by continuously warm days: now we were up to 2 weeks ahead, with growth literally exploding - by April 2nd, the shoots were up to10cm / 3 or 4 leaves on Right Bank Merlots and to 5cm / one leaf on most Médoc Cabs.
In less than a month, the vine had gone from total dormancy to vigorous growth. With a mean temperature of 9°4, March had been the warmest since '45 - and it showed. The vineyards had suddenly gone from Winter brown to Spring-like bright green. March had also continued seriously dry, with total rainfall finishing at only 42mm on the Right Bank and a tiny 28mm in the Médoc. At first, this was regarded as positive: It not only meant growers could skip a few expensive sprayings (they had to do the excoriose one but could forget the mildew one for the moment), but also - and the enormous importance of this was of course not realised at the time - it made the vines' roots go down to find more permanent deeper water reserves.
However, there was a negative side to this Spring drought which now starved the vine into producing (1) a small to average "sortie" (number of embryo bunches), and (2) "fenêtres" (gaps between bunches). These now dashed the hopes of the large harvest everyone had been hoping for to compensate for the low yields of 2002. Surprisingly, given that this was now a "vin de mars" year (embryo bunches visible before the end of March), frost was not much talked about. Everyone seemed to just laugh it off: "We'll worry about the '03s once we have worked out how to bring the '02s to market". Also, the Full April Moon was late (16th) meaning less risk. But then the early mornings started to get colder from April 3rd with the arrival of a chill North-Easterly that was to stay for 5 days. The morning of 7th saw some pots burning in the vineyards, and on 8th, it went down to -0°5C in parts of the Médoc, -1°C on the Right Bank and -1°5C in Léognan and Sauternes. This was not quite enough to do much damage, but a few shoots got frizzled here and there. More importantly, it was now talked about for the first time - but never implemented - as a factor in the 2002s pricing: What if 2002 and 2003 were to be two consecutive half crops? In the end, they would be, but from too much heat not from too much cold!
This cold patch slowed down the vegetation but not for long. After some light showers April 10th-11th, the Southerlies returned with their 23-24° afternoons, and growth continued, slowly at first, then very fast after the storms of April 26th-30th. The vineyard turned fast from dull pale to vigorous dark green again, and by 30th, the Merlots were at "boutons floraux séparés" stage (embryo grapes separating from each other in the bunch), so the 2 week advance had been maintained, in spite of the cold snap and the continued drought (April totalled a paltry 18mm in the Northern Médoc)..Early years have a better chance of being good years than late ones and growers were generally optimistic. We could not know at the time that this earliness was to put the beginning of the harvest right in the hottest part of August, and later we wished it had not been so early.
Early Summer and Flowering
May continued dry (Rainfall was 43% below average at Mérignac), and most of it fell early on, over the 3 days 5th-7th . There was another sudden surge of growth, but this time we were getting dangerously close to the flowering, and even if the cool nights calmed things down a bit, such a pre-flowering spurt was exactly what had caused all the coulure on the 2002.
The first flowers were observed over the weekend of May 17th - still 2 weeks early. Then two things happened: (1) The flowering got spun out, especially on the Merlots, first by the stormy day of 19th, then by the very cold subsequent mornings. The Merlots didn't finish until May 28th, and the Cabs until June 6th; (2) For the second year running, there was coulure on the bunches that were in full flower during the period of maximum irregular weather: a spike of heat up to 31°C on 22nd-23rd, followed by 2 days of light rain on 24th-25th, 3 very cool nights down to 8°C on 25th-27th and finally a sudden burst of 27-31°C heat on 27th-31st. These are less than ideal flowering conditions and it quickly became clear that some damage (although less than last year) had been done, once again especially to the old vine Merlots, in the form of both coulure and millerandage. These were to be precisely the grapes that would suffer the most from the heat stress later and they didn't need this problem now.
High Summer: June and July
This so far fairly routine if very early year now entered a phase that was to set it apart from all other years in the recorded history of Bordeaux. For 92 days from May 28th right the way through to August 28th, Bordeaux experienced the most unrelenting and totally exceptional heat-wave of all times. Many previous heat records were to remain unbeaten for June (number of days over 30°C: 15 in '76, "only" 10 in '03) and for July (number of days over 30°C: 17 in '49, "only" 11 in '03; average maximum temperature: 29.8°C in '49, "only" 27°8C in '03; maximum single day: 38°8C in '90, "only" 37°9C in '03). But August pulverised them all: The average maximum daytime August temperature in '03 was 32°1C (previous record 29°5C in '91); maximum single day 40°7C (previous record 38°8C in '98); number of days over 30°C: 20 (previous record 15 in '90), and of course the major title of the year: those stifling 11 consecutive days over 35°C August 3rd-13th. As elsewhere in France, that should have dealt the final blow, as it did to all those old people who tragically couldn't cope with it. But strangely, only parts of the Bordeaux vineyard shrivelled and collapsed under the stress. The rest of the vineyard continued to look dark green and flourishing all through, and went on to enjoy the showers and the fresher weather that followed, and above all, those halcyon days of September that could put the final touches to the ripening of the skins and pips long after the sugar and acidity had arrived at apparent ripeness.
All over France, we heard of how the vines stopped functioning, the grapes shut down and shrivelled, and then had to be picked in a hurry during the searing heat of August. Of course all this happened in Bordeaux too, but to a much lesser extent. First of all, with its Atlantic climate, Bordeaux's maximum temperatures were more moderate than further East. On only 4 out of the 92-day heat-wave did the wind come from the dry East: On all the other 88 days, the weather came in from the West, with at least some moisture in it, off the Atlantic. This immediately sets this '03 heat-wave apart from the blast of hot drying air we got from the East in '76, '90 and the end-of-season '86.
In addition, we have already seen how the Winter rainfall was sufficient to maintain moisture deep in the soil, and even Bordeaux's low rainfall figures for February - May made other regions envious. And now, the rainfall for these 3 searingly hot Summer months, even though most of it came in the form of storms that tend to run off quickly, was perfectly normal: Depending on the area, June saw 40-75mm (the Mérignac average is 68mm), July 35-90mm (average 54mm), and August 27-61mm (average 63mm). This was in marked contrast to the total drought of many other parts of Europe and it also clearly sets the vintage apart from other similarly hot years such as '47 when the heat had been totally dry, with no Summer rainfall at all to alleviate it.
Finally, the heat-wave was very progressive rather than in the form of a sudden burst of searing heat such as had moulded vintages like '90 into their very roasted, sometimes almost overblown character. Even for that growing band of people who de-leaf in June, the grapes, already not over-abundant, got used to being exposed to the sun and their skins hardened to protect them, as did the leaves, which tended to grow sideways at this time rather than vertically.
This progressive June heat, followed by a hot but quite wet July, really set the vineyard up to deal with the August roasting. The ultra-hot days of June 11th-15th also re-consolidated a lot of the disparity that had been created at the flowering and the bunches started to look more uniform, the Merlot ones still very long and loose, this impression of looseness being enhanced by the small size of the grapes. The Cabernets were also long but much tighter. Then the 35°-38.5°C temperatures of June 20th-22nd stopped the TGV running and were heralded as "August in June". Extreme heat at the beginning rather than at the end of the ripening cycle can fix unwelcome green tannins into the grapes' skins as it had done in '98. Little did we know at the time that this was just a prelude to a far hotter August, but it was enough to restrain growers from doing any more de-leafing.
July, with its 3 series of rain, ended up marginally less hot than June with the mean temperature at 22°2C vs June's figure of 23°1C. But the thermometer still got up to 37°9C on 13th and the month ended up a full 2°2C over the average.
All this heat was bound to provoke some sort of violent counter-reaction from the weather. So it was no real surprise when we got two cataclysmic storms on the evenings of June 24th and July 15th. Normally such storms are very local, but these two created enormous widespread damage. Both originated from a temporary retreat of the Summer high pressure system, which otherwise was situated permanently over N Europe for the whole Summer (affectionately dubbed "la patate" by French meteorologists), fending off any trans-Atlantic depressions. On these two occasions, the "patate" temporarily weakened allowing cool, moist North Atlantic air to suddenly flow into the region, violently confronting all this extremely warm low-level air and convecting it into towering cumulus clouds that moved slowly inland SW to NE, dumping deluges of rain and, worst still, copious quantities of hail.
The first one, June 24th, the third night of Vinexpo, formed over S Graves and moved quickly NW across the Graves to Portets, crossed the Garonne, pummelled the Premières Côtes and Entre-Deux-Mers around Langoiran-Capian-Créon, crossed the Dordogne at Branne, dumped more on Vignonet and St Christophe in St Emilion, before expiring in Puisseguin. 160km/h winds blew down whole rows of vines, then hail ripped through them. In this maelstrom - this was no ordinary localised hail-storm - a whopping 6000 hectares of vines were severely damaged in the Graves, Entre-Deux-Mers and lower St Emilion. Vinexpo party-goers in tuxes and long dresses had to evacuate marquees and manhandle trees off the roads to get home that night, and over 1.3 million cases of potential wine were lost.
The second one, July 15th, was just as strong, with 158km/h winds and lashings of rain and hail, but its similar SW-NE trajectory this time took it slightly further North than the first one, and most of its track was over forest, from which it emerged at Moulis, crossed the Gironde and hammered Blaye. Consequently "only" 2700 hectares of vines were damaged, mostly to a lesser extent than the previous one. But, in an already diminishing vintage, there went another 350,000 more cases.
High Summer: August
There has never ever been more heat than this. From August 2nd to 13th, every day saw 35°C+ temperatures and the all-time Mérignac record was set on 4th at 40°8C, and that day further inland at Montauban, it hit 41°8C. Up until now, it had been hot but there had been moisture in the air (There had even been a mildew scare in July), but from July 28th to August 16th, not a drop of rain fell nor was there any dew until it suddenly cooled on 14th . Yet amazingly, the South-West of France had no forest fires, as in the South-East, we were never forbidden to wash the car or water the lawn as was the case elsewhere, and around Bordeaux there was comparatively low mortality amongst older people. Similarly, with a few exceptions on lighter soils such as Western Pomerol and the plateau of St Emilion; the vine remained dark green throughout, did not lose its leaves and only really shut down from 6th-13th August. Only the exposed grapes suffered and shrivelled, and even then not all, the remainder concentrating down to almost no juice - just pips and skin - but still just functioning. Those who had de-leafed too energetically lost a lot of grapes now, whilst those who de-leafed to only 50cm or so, or whose rows of vines ran E-W rather than N-S - thus avoiding violent exposure of the bunches to the afternoon sun - could claim that they had reduced dehydration of the vine (because the leaves are the vehicles of evaporation)
During all this, the véraison had happened, swiftly and effectively, from July 25th. It was finished during the extreme heat, around August 7th for the warmer soils, a few days later for the cooler vineyards. The rapidity of this véraison caught up a lot of the irregularity that had been occasioned in the bunches during the spun-out flowering, and by mid month, the long bunches of tiny grapes that had survived were looking tired but good, whilst the ones that hadn't made it were already shrivelled and falling off.
The harvest
In the final days of the early August super-heat-wave, it was time to harvest many of the Pessac-Léognan whites. They had already lost much of their juice to the heat, the sugars were very high, and any further delay would not only bring further evaporation but any aromas that were left would be totally blown away. Accordingly, in a much-publicised move, Château L'Hospital was reportedly the first with a small pick on Tuesday August 12th, and this was followed by an even more publicised total pick of the Haut-Brion and Laville Haut-Brion whites on the Wednesday and Thursday, followed quickly by the André Lurton estates.
This was the earliest start by far, including 1990, of any vintage since 1893. Many others would have picked their whites now, but could not get their personnel back quickly from Summer holidays. So the grapes had to stay on the vines and run the risk of drying out even more.
This is where they mostly got lucky, when, from August 14th, the daytime temperature dropped from over 35°C to a "mere" 28°C., and over the weekend of 16th-17th there were a few gentle thundery showers - the same system that brought flash-flooding and gales in the Languedoc and South-East, but had not yet worked up much strength when it had gone over Bordeaux. A few mm of much-needed non-violent rain fell on the whole region, more in the SW of the appellation than on the Right Bank. But the Right Bank caught the situation up nicely in a second round of much heavier but just as gentle thundery rain overnight on Tuesday 19th, when it received 35mm whilst other regions only got 20. With the vines nicely refreshed and the grapes reviving from their wrinkled torpour, the rest of the Pessac-Léognan whites were harvested during this cooler, breezy week of August 18th. Then, a third small 4-5mm shower on Sunday 24th was exactly what was needed to set off the main body of the Bordeaux and Graves' white harvest during the sunny week of 25th, most finishing in ideal conditions around September 3rd, after a fourth little round of gentle thunder showers (5-20mm) over the weekend of 30th.
It was just now, September 3-4th, that the earlier ripening red wine areas such as Pessac-Léognan, St Emilion and Pomerol started their Merlot harvest, first under bright sunny conditions 3rd-5th, then between the showers (fortunately the really heavy ones fell at night) 6th-10th. Most of these 3 regions had finished their Merlots by 12th, in a hurry towards the end because those 5 days of 20-35mm rainfall had provoked some outbreaks of grey rot, about which there was a lot of concern, since it could spread very quickly indeed during the long period of warm fine weather that was now being regularly forecast for the following days.
That growers began to count on this fine weather during the showery days of September 6th-10th was now to become a determining feature of the vintage. If there had been rain in the forecast, all remaining Merlots and probably a good proportion of the Cabs would have been picked right away. The sugars were very high, the acidities abnormally low; only the pips and the wood of the grape needed to ripen. Better to have slightly green grapes than rotten ones, and the resulting qualities would have been as mixed as those who had already harvested .
So, of those Merlots still to be picked, mainly the Médoc ones, the better Bordeaux and Côtes ones, and the Right Bank ones on "argilo-calcaire" soils, most now had the confidence to put off harvesting until the following Monday September 15th. Others who had more grapes that were fragile, dried-out or tasteless (a very frequent comment at this time) quickly picked now over the fine weekend of 13th-14th, marginally before total ripeness, and, as a result, missed out on the fabulous, most perfect September conditions that were to come, as the "patate" high-pressure system ballooned out once more over Northern and Central Europe for two whole weeks, bringing dry Easterly and Southerly breezes into France.
Most French and half of Bordeaux's vineyards had finished their harvest by now, and could only watch with dismay what they were missing. Bordeaux's best was still on the vine and could now be harvested at leisure right the way through to the end of the month. Only two light showers sprinkled the region on 21st-22nd September (4-7mm) and on 27th-28th (1-2mm) and September's total rainfall ended up one of the tiniest of all time: 32mm (The average is 85). And the temperatures ended up very high too, with the mean temperature at 1°4 over the average and the hottest day (on 20th) at a full 32°C, very hot for September in Bordeaux. If it had been wet, such weather would have been a disaster, but, like this, with a renewed dry late-season heat-wave, it was just like in '86 - except that in '86, the grapes went into it all swollen by the late Summer rain and needed to dry out. This time, they were all concentrated already, sometimes a little shrivelled, sometimes taking time for the grape-wood to ripen perfectly, but basically, it was just a question of allowing the finishing touches to be applied to the ripening, and with the forecast so serene, no-one needed to be hurried into any untimely decisions. Anyway, if anyone was going to panic, they would have done so back in late August / early September, not now.
These later harvesters chewed at their nails over the weekend of 13th-14th September. Supposing they had miscalculated? What if the grapes had been ripe after all last week? And, above all, what if the weather forecasts were wrong? With such a small yield hanging on the vines already, it was a tough call, especially for those estate managers who are responsible to shareholders and owning families, with the consultant oenologists almost universally advising earlier harvesting. They were going for broke in the face of nearly all professional advice.
The week of 15th September dawned bright and clear, with a pleasant drying East wind. The vines were all dark green again, having been restored by the rains of 6th-10th, and the skins of the more wrinkled grapes began to tighten once more under the pulp's pressure from within. Most of the Merlots from the Médoc, from the later-harvested "argilo-calcaire" St Emilion and Bordeaux-et-Côtes vineyards were picked during this glorious week of 28-32°C days and cool nights, the hot days rounding off the ripening, the cool nights returning a semblance of acidity to the juice.
On the Sunday night September 21st, the remains of hurricane Isabel arrived on our shores, but it was by then pretty much spent, and the total rainfall that night was only 4mm on the Right Bank and 7 on the Left. This put the final touches to the ripening of the Cabs (Franc on the Right Bank, Sauvignon on the Left) so that harvesting could commence - or in some cases continue - during the following 7 days of sunny if cooler conditions, only interrupted by very light showers 28th and 30th September that did absolutely no harm at all. Most of this week's harvest was first class, only compromised here and there by a few shrivelled bunches.
Some lingered on for a few more days into October up to Friday 3rd, in damper and more stormy weather, not looking so good as the previous week but with generally better results for these final parcels than if they had not waited. But the great bulk of Bordeaux's Cabs were in by 27th and immediately the juice displayed all those dark colours and strong structures that had been missing from the earlier Merlots and whites. At this time, it was thought that this was clearly going to be more of a Cabernet vintage than a Merlot one. It had happened just in time. It had been an early vintage, but it had finished much later than originally planned at the véraison, and nowhere near as early as Burgundy which had finished by the end of August, and it was extremely fortunate not to have gone on any longer: For right then the heavens opened and October ended up incredibly wet at 155mm (the average is 87). Nobody could quite believe their luck, and maybe because the mass of early harvesters were cursing theirs, those who had been able to harvest later could not quite come to terms with the fact that they had possibly produced something rather special. Even at the time of writing, they are only just beginning to admit that at least some of what they have produced is truly outstanding, and then largely under the persuasion of outsiders who had not witnessed all the difficulties of the pre-harvest and who just figured from afar that a hot Summer = automatically great wine.
Vinification
In a very short time, Bordeaux has made tremendous progress in harvest-reception and vinification equipment, right across the board. Ripeness analyses are now extremely detailed and untold efforts are made to ensure that the grapes arrive at the cellar in perfect condition (bunches brought in in crush-proof boxes, special hoppers to prevent pre-cellar crushing); introduction into the de-stemmer-crusher above the tanks by gravity in order to avoid pip-crushing screw pumping; double vibrating sorting-tables separated by the de-stemmer to ensure that no even minor imperfect grapes get through; boosted cooling systems that will allow several days of pre-fermentation maceration whatever the ambient temperature (very important this year in order to gently extract the laggard fruit flavours of the year); and above all, re-building of the cellar into smaller tanks, often returning to concrete or wood (less brutal thermic shock than sta