There is certainly some from the paint and the tires (mostly oil right?). I wonder what else?
EDIT: The initial text was stating that a bike pollutes more then a car which is obvious shit. This is just my english at its finest. I'm sorry for this.
Following on from that, the pollution generated from day to day upkeep will be less as well (when was the last time you had to change the oil on your bicycle?)
Lastly, the energy required to power the bike (the energy that you provide), is a bit of a sunk cost, you need to eat to stay alive after all, and while you _might_ need a little more, it's still not going to be equivalent to driving your car to work and back every day.
First of all, your metabolism isn't going to stop (assuming you're still alive), so it's not really fair to say that those calories you burned wouldn't have been burned if you hadn't biked. I exercise a few times every week. If I bike less, I just have to do more exercise, but i burn about the same number of calories.
Secondly, most bike trips are short and relatively low speed. Biking encourages you to explore your local area. Lower speeds mean lower wind resistances, and shorter trips obviously make the mileage considerations a lot less important. I really don't think I'd be willing to travel more than a few miles by mike, since I don't want to spend hours in transit.
Overall whilst a bike isn't carbon neutral to make, it'll produce less over it's life time.
Let's guess that a car weighs about 3000 lbs, and then let's guess that a bike weighs about 20 (obviously there is large variance). Mining or producing all those resources produces pollution, we can even generalize it to be similar levels of pollution. The 3000 lbs. of raw materials times the pollutant level will, by far, be the dominating factor in the 'how much pollution did creating this thing create' equation.
However, there's a slightly more interesting analysis to be had if you consider that car fuel is way more efficient than bike fuel (human calories). So maybe, just maybe, a sufficiently poor diet can cause riding a bike to work to be worse for the environment than driving.
I have a "Giant" from 1992 and I am not a serious bike rider but I believe I've purchased more tires for the bike than for my car, despite commuting probably 100x further in my car. Bikes are a serious headache to maintain compared to a decent car.
If someone could design a bicycle that lasted as long as a car for normal users, that required as little maintenance as a car for the first decade or so, that was as reliable as a car, that blew out tires as rarely as a car, then it would be one heck of a bike... unfortunately due to lack of economy of scale and having almost nothing in common with conventional bike design, it would probably cost almost as much as a car...
Look at nations that have cycling as a way of life. "Mama fiets" (apologies for potential misuse of Dutch language) are robust bikes for carrying children and shopping around.
There are puncture resisting tapes that can be applied to tyres.
Most people don't need fixed wheels, nor 20 gear derailurs, but would do well with 5 well chosen gears.
Aluminum frames (with modern welding and tube shapes) are light and robust. And probably a good use of recycled aluminum.
And people tend not to educate themselves about cycling. Having the seat post at the right hight makes a huge difference. Using the balls of the toe on the pedal and not the arch of the foot makes a huge difference.
Many people would be served well by a £500 bike and a book about cycling. Include a basket and some panniers and mud guards, with a can of spray lubrication and a few spare tyres and replacement chains.
I agree that cycle shops usually are not set up for this, and that the image of flourescent lycra is off-putting to many people; and that it would be great if commuting by bike was more accessible.
It's definitely possible for your bike to need way less maintenance than a car and to provide very few headaches.
In your case, I recommend that you buy very good tires. Most bikes have cheap tires, which dry rot. Some of my bikes have tires that are 6-10 years old and still working fine. They also cost $40 apiece back then.
In general, most bikes are designed for the recreational "joe 6 pack" market and are of poor quality. If you expect a bike to work well and last a long time, you have to pay extra.
Car vs. bike discussion may have (some) sense when it comes to commuting distances, adverse weather conditions etc.
I have a super minimalistic garden and grow some of my own salad vegetables in the summer but I'm under no illusions and 99% of my food calories come indirectly from crude oil and natgas.
My car doesn't burn biodiesel and isn't made of soy. The store is illuminated and HVAC by natgas and coal. The delivery truck from the warehouse doesn't burn biodiesel and isn't made of soy. The warehouse burns lots of coal for its HVAC and lights. The trucks delivering to the warehouse don't run on biodiesel and aren't made of soy. Assuming produce and not "processed food-like substances" the farm uses diesel to run the tractor and other gear, some coal powered electricity in the greenhouse, fertilizers made of phosphates dug out of the ground (what to do when they run out?) and nitrates made by the haber process outta lots of natgas. Also there may or may not be herbicides / pesticides / fertilizers made out of crude oil.
If your food is cheaper per "gallon" than diesel, its at least theoretically possible to make it with biodiesel and be net energy positive. If it takes 20 gallons of diesel equivalent environmental damage to make a gallon of walnut oil, then if you poured walnut oil into a biodiesel tank the system would rapidly run dry and halt.
Computer science types have a stereotype where the "halting problem" can't be solved for them so it can't be solved anywhere, although energy calculations are actually pretty simple (yes yes I know totally different type of halting its a bad joke). If (Energy Returned) is less than (Energy Invested) then the only way to keep pumping food out is to burn crude oil to make up the difference.
Depending what axe you have to grind, EROEI of food varies from 0.1 to 1. The truth is probably within that range, perhaps 0.3. A pound of fat (animal hydrocarbon, basically) is worth 3500 or so calories so the environment is better off if you burn two pounds of gasoline than if you eat a pound of food (aka equivalent of 3 pounds of diesel burned to make it) and bike the same distance.
One big problem is the EROEI of food varies so much. Kobe beef flown in by air from Japan as a bicyclist fuel is likely to be vastly worse than driving a typical car, yet eating home grown vegetables is likely to be better than driving a V8 crew cab pickup truck. No debate there. Are you better off eating a little more factory farmed chicken on factory farmed rice fried in chemical plant produced psuedo-oil, or burning gasoline, well there's a debatable area.
You'd do best to google for EROEI and related terms. Warning that all reports in the field are highly politicized, so you need to average the archived stuff on theoildrum showing a 0.1 with some multinational megacorp astroturfed greenwashing campaign claiming a 10.0 EROEI.
There are no unbiased academic textbooks I'm aware of. Its all "The Peak Oil Myth Debunked" or the "Partys over" book and "Crude awakening" movie and there is nothing in between.
Generally the pessimistic treatments do a good job of science and engineering and math and do a bad job of politics, economics, PR and emotional response. And the optimistic treatments do the obvious swapped topics. So if you critically read both and pay attention appropriately, you'll come out with the best of both (try not to mess up and end up with the worst of both sides LOL)
For just one book, you could do worse than read Kunstler's emergency book. Not because its perfect or even because I agree with all of it, but because it is extremely content dense WRT listing the many inter-related systemic problems. It makes at least passing mention of almost all issues, and at least one sides thoughts about it, regardless if you agree or not. I don't think the "other" side has anything near as information dense, that I know of.
3 hours on a bike at a somewhat moderate pace isn't going to be more than 2000 kcal. Of course, animal fat is like 10 times less efficient than directly eating the vegetable calories.
I guess an electric car that recycles well might end up beating the bicycle though.
I got 160K in 16 years out of my last car. It cost about $20K. I figured its total lifetime cost including gas and what little maintenance it needed at thirty five grand. So figure about 20 cents per mile. That's cheaper than average, then again it was more of a beater of an old car that most would tolerate. Mechanically fine, but cosmetically a mess. So yes twenty cents per mile is not totally ridiculous.
Being a purely recreational biker I suspect I've only got 1K or so miles on my "Giant" bike from 1992. It was about $300 but the way tires decay (why so much faster than car?) and general minor repairs I'm sure I'm in for $500 by now. So I'm paying 50 cents per mile, or about 2.5 times more than my commuter car costs the environment.
Worst of all, being a recreational bike, I've caused $500 worth of environmental destruction just for fun when I could have meditated or bird watched or gone hiking. And if you get rid of the recreational bikers, that's a large market segment so costs and enviro-damage will increase. Aside from the obvious win of get rid of recreational bikers, well, they'll just drive to the mall with their new spare time or whatever.
I'm sure I could game the numbers and if I had a recreational car and a commuter bike the numbers could be made to look different. One problem with a commuter bike is the weather only cooperates about 1/3 of the year, and my car goes easily six times faster than my bike, making the commute punishingly impossible.