Unless unprecedented carbon capture systems are discovered petrocarbon-based energy systems are not defensible in the medium to long term regardless of what flavor of market fuckery you try to bring to bear. The greenhouse effect isn't concerned with or impacted by geopolitics.

You are only incentivized to reduce leaks if the profit from selling the additional natural gas is larger than the cost of reducing the leaks. It is difficult to track down all the leaks because you cannot see a leak with your eyes (like you can see oil leaking from a pipeline). The key difference is that it is much harder to stop a gas from leaking than it is to stop a liquid like oil from leaking.

Most of what I'm saying is coming from a lecture by Prof. Howarth at Cornell this past Fall.

His full slides are linked at [1]. His second lecture in the series [2] suggests upstream emissions from a well are about 1.3%. That's not enough to justify significant investment in capture for production increases, but it's enough to do enormous damage to the environment. The first set of slides explains that damage reasonably well, but the TL;DR is that Methane screws us far more in the short term, and all the timelines that matter at this point are short term.

I don't have any information on biofuel production. I think it's an open question whether it's useful for grid-scale energy production -- again, gas turbines are wonderful base load generators -- but I'd be extremely hesitant to support it's use in transportation.

Oversea shipping is a challenging problem that may be served well by gas as a stop-gap, but long term it should move to compact nuclear reactors which use non-weaponizable fuel elements or new energy dense storage methods like lithium air batteries.

For ground transportation, it's really hard to justify anything other than direct electric vehicles at this point. The energy density of natural gas isn't that great relative to current Li-Ion packs when you consider tank weight, and any pure ICE vehicle has to deal with large braking losses, which EVs can recover trivially. The various flavors of hybrids are a decent solution, but I'm a huge proponent of the pure EV model with parked EVs doubling as a grid-scale storage solution.

[1]: https://drive.google.com/file/d/0B1ocJG5BVzw0TmRmSUs4TjJRdVk... [2]: https://drive.google.com/file/d/0B1ocJG5BVzw0aUVUQ2tEcXNwTVU...