I'll suggest that literacies, scientific and others, are something like physical fitness. If you only know how to lift weights, or run, or exercise in general, you're not physically fit. You have to do the weight lifting, running, and other exercises to be physically fit. So it goes with literacies. If you have a knowledge to do something, whether that's reading, math, or bringing a scientific way of looking at scientific problems, but you don't actually do so on a fairly routine basis, you're not really literate at those things. You could become so, same as I will become physically fit again. But I haven't been exercising regularly, so I'm not physically fit at the moment.
Something mentioned in the comments I invited was there being a progression to scientific literacy. I'd never thought about it before, but that does make perfect sense. In physical fitness, there's a progression from getting out a couple of times a week and doing 20-30 minutes of something constructive, to 3-4 times a week at 30 minutes, to ... well quite a range. So it goes for scientific literacy -- a couple times a week reading and thinking about it, several times a week and applied to more challenging materials, and so on.
Related to that, though, is that the bulk of the population does not need to be at professional levels of activity to be 'fit', or 'literate'. I'll be in pretty good shape when I'm exercising my 5 hours or so a week. For running, that'll be about, oh,
20-25 miles per week with other activity, like weights, thrown in. But a competitive runner would be going 70-150 miles per week. I won't be doing that. And I'll be running at, say, 10 minutes per mile, as opposed to the 6 minutes of the elites. Still, I'll be fit -- even if not ready to challenge professionals to a head to head race. The scientifically literate person won't be ready to publish in the professional literature, but they can certainly read reports about it, and even (at higher levels of literacy) read the original papers themselves and gain from the reading.
A number of people mentioned particular items that should be known. Sometimes with the mention that there was too much, any more, to be known for possibly anybody to be truly literate. I'll disagree, really on both sides. Going to literature, for the moment, 200 years ago there were rather few books being published. At that time, one might have a sense that being literate meant that you'd read some notable fraction of the books being written (or that had been written). Then comes modern publishing and far too many books published each year for anybody to read any significant fraction of them. Did a reading rate that was literate 200 years ago become illiterate today? Or was that just not a good way of defining literacy? I opt for the latter. It has probably never been the case that anybody could know all that there was in science, and certainly not any time in a very long time.
My own take is that scientific literacy is less a matter of the fraction of all knowledge that you know, and more a matter of how you react to new knowledge that is offered up to you. The scientifically literate person, for example, may well not know that ozone is also a greenhouse gas. But, on encountering a claim that it is, be able to tell that it is important to the argument at hand that it is, and know how (and carry out) to check as to whether it is.
In this vein, I agree strongly with the folks suggesting that one of the skills of a scientifically literate person is that they can (and, I'll add, do) evaluate whether the source they're looking at is scientific or not. Things like my 'weeding sources' suggestions become second nature to the scientifically literate people. Acquiring new knowledge is no help if that new material is false. In that vein, as I've commented elsewhere with John Mashey about his K index (do take a look at it in the comments already linked to), I think there's such a thing as negative knowledge -- thinking things that are not true. For example, thinking that the earth is flat, or the sun circles the earth are negative knowledge. That's much worse, to my mind, that merely not knowing that the earth is round and circles the sun.
Where, then, does knowing science factoids come in? Certainly all the scientifically literate people I know (and I do consider that I know several nonscientists who are scientifically literate) do know a fair number of science factoids. My feeling is that this is a result of being scientifically literate, not the cause. For fitness example, right now I'd probably cover a mile in about 15 minutes, mostly walking. As I become more fit, I'll cover that mile faster, eventually getting somewhere around 6 minutes. But running the mile faster is not fitness -- I wouldn't necessarily be more fit if I ran 5:51 instead of 6:20. I might actually be in best fitness at a time when I was running the mile in 6:40 -- better endurance, strength for trail running, better core strength, etc..
The process of evaluating a report about some scientific work is, for scientific literacy, then, like going for a training run is for physical fitness. The training runs are where you improve your physical fitness -- build and train muscles, your cardiovascular system, and so on. The evaluation process -- which will involve learning the meaning of terms, some elementary facts, connections between facts in that field -- is the training run for scientific literacy. After you've done the training, you know facts. But they're the byproduct, same as a better mile time is for being more physically fit.
There is a serious practical problem in this, and maybe some teachers (hint to my sisters) will add some comments on that. Namely, state and local curriculum requirements are heavily oriented towards memorizing facts, rather than 'how to do science', 'how to evaluate scientific claims', and the like. Teachers who simply started running their classes in the direction of my notion of scientific literacy would probably be fired for failure to do their job (as the states and local school boards see it).
Still, maybe teachers can chime in with their observations of either what the literacy is, or how to implement ideas like this in the classroom.