Best inflation pressure of car tire in the rain- higher, same, lower than normal?

If I am driving a car in the rain, and want to increase the available traction, should I:

1. Increase pressure in the tires
2. Decrease pressure
3. Leave the pressure set to optimum dry pressure
4. Increase or decrease just the front or back tires

Does the answer change if the surface has no standing water (i.e. is merely wet)?

-

Increased tire pressure certainly reduces the tendency of the vehicle to hydroplane, that is, to ride on a layer of water. That's not the same thing as contact friction with the road. Once the tire is actually contacting the road and the pressure of the water layer pushing up is not so significant, then pressure does not appreciably affect traction.

-

This link provides a detailed with pictures reason why underinflated tires have worse control during hydroplaning. It also describes a road trial under the two conditions.

-

The tire pressure will not affect the traction in a noticeable way. The reason the car moves forward is because of friction from the ground, this friction is generated by a downward force (of the spinning tire), which has the effect of pushing the car forward. This downward force is influenced primarily by the mass of the car, and not the tire pressure; which is why they sell special tires with better traction. The traction depends not on the pressure, but rather on the material and geometry of the tire surface.

-
Increasing pressure decreases the size of the contact patch, which changes the geometry of the tire surface. It should cause the ground pressure to increase because the same amount of weight will be supported by a smaller area. Doesn't the increase in ground pressure mean that the tire can push more water out of the way, thus reducing hydro-dynamic effects? I would expect that to improve traction. – Gabe Mar 2 '11 at 8:40
At the physics-101 level of analysis, this is correct: the traction is independent of the pressure. The static friction force (i.e., the traction) is simply $\mu_s N$, and both the coefficient of friction $\mu_s$ and the normal force $N$ are independent of pressure. Reducing the pressure increases the contact area but does not affect the traction. But rolling friction, I've always heard, is not well-modeled by this sort of analysis. How sure are you that your answer is correct? Can you supply a bit more detail? – Ted Bunn Mar 2 '11 at 14:58
This is just wrong and self-contradictory. You say it is independent of pressure but depends on geometry. The geometry changes with pressure. While driving a car the difference is not easy to feel as you rarely drive close to the limits. With a bike test it yourself, drive and brake with a pressure of 1,2 or 3 bar. – Alexander Feb 19 '12 at 12:09