I have a lot of trouble believing that the leak rate for power plants is anywhere near 2,9%. Perhaps for home distribution you might get a couple percent (I have no trouble believing that!), but NG power plants involve dramatically simpler distribution systems, much more controlled circumstances than hundreds of millions of homes withpipes in various states of disrepair snaking through walls and floors, and people with their pilot lights out spewing gas, people leaking gas as they fire up their stoves, etc etc.
I'll need to dig through some of the papers when I get a chance, to see what they actually say. But I seriously, seriously doubt any meaningful fraction of the leaks come from peakers and the distribution infrastructure leading up to them.
Furthermore, let's just pretend it's true. Let's pretend that peaker warming impact is literally doubled. So? Peaker warming impact is almost nothing compared to other sources. Almost nothing times 2... let's see.... multiply by the nothing... carry the nothing...
Your article (from "Thenarwhal.ca") doesn't load for me, but from the title, it does not sound like it's about peakers. Coal isn't being replaced by peakers, it's being replaced by combined cycle baseload NG plants.
And I'll reiterate: they can't "go" because we literally do not currently have a realistic, cost-effective system to replace them in terms of backup power for multi-day geographically-broad wind-solar lulls, which legitimately do occur far too statistically commonly to ignore. Maybe some day we will, but that day is not today.
Try pricing amortizing, say, 5 days of storage. Tesla's large projects are $350/kWh installed. That's $42k/kW/5d. Amorized over the 15-year lifespan at 5% interest, $332/mo/kW. Do you find that acceptable? And we're only talking a 5-day lull. Bad weather patterns can lock in longer than that. The worst case is a major volcanic eruption, which can dim the whole globe dramatically for many months on end. Laki's last eruption for example froze the Mississippi at New Orleans. Writers wrote of burning-glasses (magnifyng glasses for starting fires) being unable to ignite dry brush.
But that's an extreme case. Simple lulls, however, are not.