Since Michelin's return to Formula One in 2000 the weather during Grand Prix weekends has been described not as 'cold' or 'overcast' or 'wet' or 'boiling', but as 'Bridgestone' or 'Michelin' weather. Of course, the one or other tyre company has unexpectedly won in conditions better suited to the other (when other factors may well have come into play) but, generally, Bridgestone has outperformed the French product when the going has been cool(er) or wet, whilst Michelin has shone in hotter, drier climes.

That being so, despite the usual insistence of both tyre companies that their compounds are best under any conditions, and with indications being that these trends will continue into 2004, surely the major changes wrought to the 2004 calendar - both in terms of quantity of Grands Prix and their placement relative to last year - will bring with them changes to F1's weather patterns? In short, does Brazil's shifting from the end of March to late October bring with it 'Michelin' or 'Bridgestone' weather? Could the re-inclusion of Spa favour Bridgestone? What are the expected effects of a June race in the United States, or a race in Bahrain in April?

Formula One has, of late, visited southern or equatorial regions in March/April, and then stayed in the North thereafter. Not anymore. True, Australia/Malaysia are in their standard slots, but Brazil has moved from third out of 16 to last of 18 Grands Prix, whilst mid-May's Austria has gone, effectively replaced by a Bahrain race in early April. Indianapolis has been pulled forward from September to late-June, whilst the Nurburgring has also been advanced, by a full month from late June to a week after Monaco. Then China, of which F1 has no climatic experience, has joined the fray�

In fact, there are eight instances of either complete change, or, at the very least, a variance of two weeks - which, in certain regions (Nurburg, for example), is tantamount to a seismic shift. Changes are summarised in the table below:

2003                                       2004
Town          Country    Date              Town        Country      Date 
Melbourne     Australia  09/03  no change  Melbourne     Australia  07/03
Sepang        Malaysia   23/03  no change  Sepang        Malaysia   21/03
Sao Paulo     Brazil     06/04  change     Manama        Bahrain    04/04
Imola         Italy      20/04  no change  Imola         Italy      25/04
Barcelona     Spain      04/05  no change  Barcelona     Spain      09/05
Spielberg     Austria    18/05  change     Monte Carlo   Monaco     23/05
Monte Carlo   Monaco     01/06  change     Nurburg       Germany    30/05
Montreal      Canada     15/06  no change  Montreal      Canada     13/06
Nurburg       Germany    29/06  change     Indianapolis  US         20/06
Nevers        France     06/07  no change  Nevers        France     04/07
Silverstone   England    20/07  no change  Silverstone   England    11/07
Hockenheim    Germany    03/08  no change  Hockenheim    Germany    25/07
Budapest      Hungary    24/08  change     Budapest      Hungary    15/08
----------    -------    -----  change     Spa           Belgium    29/08
Monza         Italy      14/09  no change  Monza         Italy      12/09
Indianapolis  US         28/09  change     Shanghai      China      26/09
Suzuka        Japan      12/10  no change  Suzuka        Japan      10/10
----------    -------    -----  change     Sao Paulo     Brazil     24/10

Note: changes only indicated where 2004 dates over last year's equivalents vary by at least two weeks; or where country/region changes.

Quite clearly then, weather effects are to be expected, but can these be quantified? Obviously, weather can never be forecast with total confidence (ask Williams about Monaco 1997), but trends can be analysed and deductions drawn. By comparing temperatures and rainfalls per season it is possible to predict a trend. Does it favour the Japanese or French company?

So, taking 2004's 18 circuits, what are the temperature and rainfall expectations in comparison with the season past?

Basis:

1. Daily temperature highs and monthly average rainfalls (as quoted by www.weather.com or, if not available from that source, from other internet-based sources) based upon dates listed above have been utilised. Where an event falls within the middle (plus or minus one weekend) of a month rainfall figures have been applied 'as is'; where the event falls in the first/last weekend, a plus or minus 25% factor drawn from the previous/next month has been applied.

2. Where no direct records for a particular circuit exist, comparisons are based upon those for the closest major centre. (In brackets)

3. Average monthly temperature lows have been excluded as events are invariably scheduled for the hottest (or close to) periods, ie. F1 stages no 2004 events at 04.30am or midnight.

4. * denotes no 2003 comparisons available

5. ** denotes no 2004 comparison included

Melbourne, Australia                           Change over 2003

2003 average temperature high/rainfall actual:    24C/45mm   No change
2004  average temperature high/rainfall expected: 24C/45mm

Sepang, Malaysia (Kuala Lumpur)

2003 average temperature high/rainfall actual:    33C/231mm  No change
2004 average temperature high/rainfall expected:  33C/231mm

Manama, Bahrain - Addition/replaces Spielberg, Austria

2003 average temperature high/rainfall actual:    ---------- +24C/0mm*
2004 average temperature high/rainfall expected:  24C/0mm

Imola, Italy (Bologna) 

2003 average temperature high/rainfall actual:    18C/66mm   No change
2004 average temperature high/rainfall expected:  18C/66mm

Barcelona, Spain

2003 average temperature high/rainfall actual:    20C/44mm   No change
2004 average temperature high/rainfall expected:  20C/44mm

Spielberg, Austria

2003 average temperature high/rainfall actual:    28C/74mm   **
2004 average temperature high/rainfall expected:  ----------

Monte Carlo, Monaco

2003 average temperature high/rainfall actual:    22C/43mm   -2C/+5mm
2004 average temperature high/rainfall expected:  20C/48mm

Nurburgring, Germany (Cologne) 

2003 average temperature high/rainfall actual:    21C/90mm   -2C/-5mm
2004 average temperature high/rainfall expected:  19C/85mm

Montreal, Canada

2003 average temperature high/rainfall actual:    18C/84mm   No change
2004 average temperature high/rainfall expected:  18C/84mm

Indianapolis, United States

2003 average temperature high/rainfall actual:    23C/71mm   +6C/+34mm
2004 average temperature high/rainfall expected:  29C/105mm

Nevers, France (Bourges) 

2003 average temperature high/rainfall actual:    25C/51mm   No change
2004 average temperature high/rainfall expected:  25C/51mm

Silverstone, England (Birmingham) 

2003 average temperature high/rainfall actual:    21C/46mm   No change
2004 average temperature high/rainfall expected:  21C/46mm

Hockenheim, Germany (Heidelberg) 

2003 average temperature high/rainfall actual:    25C/70mm   No change
2004 average temperature high/rainfall expected:  25C/70mm

Budapest, Hungary

2003 average temperature high/rainfall actual:    24C/45mm   +2/+5mm
2004 average temperature high/rainfall expected:  26C/50mm

Spa, Belgium (Liege) 

2003 average temperature high/rainfall actual:    ---------- +28C/66mm*
2004 average temperature high/rainfall expected:  28C/66mm

Monza, Italy (Milan) 

2003 average temperature high/rainfall actual:    24C/74mm   No change
2004 average temperature high/rainfall expected:  24C/74mm

Shanghai, China

2003 average temperature high/rainfall actual:    ---------- +26/130mm*
2004 average temperature high/rainfall expected:  26C/130mm

Suzuka, Japan

2003 average temperature high/rainfall actual:    21C/130mm  No change
2004 average temperature high/rainfall expected:  21C/130mm

Sao Paulo, Brazil

2003 average temperature high/rainfall actual:    27C/122mm  -1C/+20mm
2004 average temperature high/rainfall expected:  26C/142mm

After all that research, can any realistic deductions be drawn? Patently not if temperature and rainfall 'changes' are indiscriminately averaged across the board. But, should 2003 actual averages be compared to expected 2004 equivalents, then a trend will emerge from which to draw certain deductions.

2003 temp. average (total highs divided by number of events):  23.4C
2003 average rainfall (ditto):                                 80.3mm

2004 expected temperature                          23.7C  (var +0.03C)
2004 expected average rainfall                     81.5mm (var +1.2mm)

Nothing conclusive there, is there?

How about comparing newcomers to existing events? Bahrain is slightly (but so close as to make no difference) above the temperature average, but well below the rainfall mean (it is reckoned to have zero moisture) - a 'Michelin' race, then? Belgium's temperature is well above 2003's average, but so, crucially, is its rainfall - it could fall either way, and probably will. That leaves Shanghai, which is above average in both measurements - meaning it could go the way of Michelin's runners on temperature or Bridgestone on rainfall. Another 'either way'�

Which brings us to Interlagos in late October instead of early-April - hardly any variance in temperature and a slight increase in precipitation. This indicates that Bridgestone should have the upper hand at the season finale in the event of expected deluges.