It's not actually that clear...
 While transportation causes harmful emissions, the food you eat and the things you own are just as significant.
We set out to find how much of a change it would make environmentally if one cycled instead of drove to work and then further expanded our findings to include the variables of diet composition and material goods ownership.
When considering greenhouse gas emissions, the picture that commonly comes to mind is one of millions of vehicles spewing out CO_{2} and other harmful products into our atmosphere. As a result, it is easy to imagine that a reduction in driving can greatly reduce our individual carbon footprint. However, this is a complicated issue and a deeper look is in order.
The problem with commuting like this  is that you eat more of this  and that may be worse than driving this.
Food may not be an obvious source of greenhouse gases, but when one thinks about the amount of energy that goes into producing food, particularly meat, and the much smaller amount of energy we get out of it, you can start to see the problem. As Albert Bartlett has observed, modern agriculture is "the use of land to convert petroleum into food"1, and it is an energyinefficient process. The energy contribution from the sun is small compared to that from the burning of fossil fuels. Tractors and combine harvesters have to be driven, food must be transported from the source to the consumers, and supermarkets have to be heated and lit. Meat production is especially heavy on the environment. Livestock emit greenhouse gases through their life processes and the energy and facilities needed to accommodate their living needs. In addition, there is a 90% loss of food energy when plants are converted to meat through animals eating it. Particularly in the developing world, carbon sinks (forests) are frequently destroyed to create poor quality farmland. All told, the feeding of 7 billion people accounts for about a third of global greenhouse gas emissions.
How does this affect our commuting choices? Well, the human body is a heat engine that converts chemical energy to mechanical work, just like a car engine, so if we cycle we will inevitably end up eating more. So the trade off is between a light vehicle (a bicycle) with inefficient energetics (in terms of mechanical work out divided the total agricultural energy input) compared to a heavy vehicle (even the lightest car is 10 times heavier than rider+bicycle) and much more efficient energetics (in terms of mechanical work out divided the total oil industry energy input). It takes very little effort to get a MJ of oil energy out of the ground compared to growing MJ of food energy.
But to come to any conclusions we need to do numbers
We gathered our numbers from various sources (shown by links in the spreadsheet) and made the following assumptions:
 The person in question weighs 70kg, is 180cm tall, and is 25 years old
 The person loses 1kg in weight due to biking to work instead of driving
 Their biking speed is 20km/h
 The commute is 10 km (one way) and is done 200 days per year
 The car is a Honda Civic (as owned by one of the authors)
On the right hand side of the spreadsheet below you will find input numbers and intermediate calculations. On the left hand side you will see the results of those calculations as kgCO_{2}e/person/year for vegans/carnivores that are active/not active and the changes between each of their CO_{2}e emissions. The legend for what the different styled cells mean is in the top right hand corner. Below the main calculations at the bottom there is a portion for you to calculate your own carbon “foodprint” and how it compares to being entirely vegan and entirely carnivorous.
Sample Calculations
First, calculate the Basal Metabolic Rate of the person using the Mifflin Equation2:
Where w is the weight of the person in kg, h is the height of the person in cm, a is the age of the person in years, f is a compensation factor for gender, and m is a multiplier that depends on the person's level of activity on a daily basis. Plugging in the values for the person above and substituting the factor for a male (f=5) and the multiplier for a lightly active person (m=1.375) will yeild:
When this is repeated for the same person who has lost 1 kg of weight due to cycling the result is a slightly lower number:
Thus it is plain that weight loss through keeping fit doesn't make a huge impact on one's intake. Now we will add some calories to the BMR of the cyclist to account for cycling to and from work.
The calories that need to be added can be found in a few steps.
First, we will find the power output needed to drive this person on the bicycle by multiplying a mechanical efficiency constant3 by the rider's weight and speed:
To add this power output to the number of calories per day, we will convert the calories per day to watts to show that food calories can be shown as a unit of power, add them, and then change them back to calories per day:
Now all that is left is to multiply by the kg/((kcal/day)(year)) for a vegan diet found from the Brighter Planet food emissions calculator4 to each daily caloric intake to find the amount of CO_{2}e emitted from eating:
At this point it looks like the cyclist emits more GHG than the driver, but we still need to add the emissions from driving.
We will calculate this using the emission values for a Honda Civic per km:
The final total emissions for each would end up being:
The spreadsheet shows a similar calculation for a carnivore.
Results
 A vegan cyclist commuter produces 300 kg less CO_{2}e than a vegan Civic driver
 A carnivorous cyclist commuter produces 600 kg MORE CO_{2}e than a carnivorous Civic driver
So cyclists can do some good, if they watch what they eat.
What about simply owning a car?
To get an idea we can divide the total amount of emissions of a country by its Gross Domestic Product (GDP), the total market value of all goods and services. In North America, that number is about 600 g of CO2e per USD; in Japan or Germany it is about 400 g per USD. If we assume that making a car involves many aspects of a nation's economy, we can estimate the GHG impact of making and owning a car.
More assumptions:
 A typical car is driven 20,000 km per year (i.e. not just commuting)
 Costs for simply owning a car are about $5000 (we assume a $20,000 Honda Civic lasts 10 years and costs $2,000 annually to insure and $1,000 annually to maintain).
Our sample calculation shows that the environmental cost of simply owning a small car like a Civic is comparable to that incurred by driving it; in each case the total is about 3 tonnes CO_{2}e per year.
To get more sophisticated than this is VERY complicated, and even after doing it, one would probably end up not learning a whole lot more. The reason is that the numbers are high; we conclude that just owning a car and leaving it in the driveway carries an annual footprint of several tonnes of CO_{2}e. This number is bigger than any involved in the short commutes we have been dealing with.
Sample Calculations
First, we find the emissions per dollar of Canada by dividing the total GHG emissions of Canada5 by its GDP6:
We can then multiply this by the approximate annual costs of owning a car to get the consumerism emissions of the car:
If the car we are considering has a yearly milage that differs significantly from the typical car milage of 20000 km/year we can divide the emissions we got above by 20000 and get a number for emissions per km.
But as we assume that our car has a typical driver we can leave the car ownership emissions as 2.5 tonnes CO_{2}e/year.
Results
If our cycle commuter owns one less car than if he/she drove to work, the numbers are much more stark:

The vegan save 3 tonnes CO_{2}e annually

The carnivore saves 2 tonnes CO_{2}e annually

The cycling vegan produces 8 tonnes less than the cycling carnivore

The cycling vegan produces 12 tonnes less than the driving carnivore
Things we have not considered
It would be very hard to quantify the environmental benefits of keeping fit by cycling and staying out of our costly (and heavy carbonfooted) medical system for as long as possible. Some may observe that we do more damage by living longer; those who think that way should take up smoking.
Conclusion
It is clear that however you "fuel" your bicycle, commuting by pedalling is MUCH better than driving IF by doing so, you own one less car. And however you get from A to B, tucking into a tofu curry is always better than having a New York steak.
Mixed message for cyclists
 1. Bartlett, A. (2004). Arithmetic, Population, and Energy. http://www.albartlett.org/presentations/arithmetic_population_energy_transcript_english.html [2]
 2. Mifflin, M.D. et al. (1990). A new predictive equation for resting energy expenditure in healthy individuals. American Journal of Clinical Nutrition, Vol 51, 241247.
 3. Wikipedia. Bicycle Performance (online). http://en.wikipedia.org/wiki/Bicycle_performance [3][April 5, 2012]
 4. Brighter Planet. CM1Diet (online). http://impact.brighterplanet.com/models/diet [4] [April 5, 2012]
 5. Environment Canada. Green House Gas Emissions Data (Online). http://www.ec.gc.ca/indicateursindicators/default.asp?lang=en&n=BFB1B3981 [5] [April 5, 2012]
 6. Statistics Canada. Real gross domestic product, expenditurebased (Online). http://www.statcan.gc.ca/tablestableaux/sumsom/l01/cst01/econ05eng.htm [6] [April 5, 2012]
© Physics and Astronomy Outreach Program at the University of British Columbia (Zendai Kashino, Chris Waltham 20120315)