Can I make a contribution to the discussion? I'm utterly clueless about the late game, but I've got a fair bit of experience of the first 100 turns
.
Scout First vs Worker First:
I want to rationalise why Scout->Worker makes sense, especially on harder difficulties. In a typical game, a Scout costs 5-10 turns, vs 15-20 for a Worker. So:
1. Worker second (rather than first) typically puts your economy approximately 10 points (eg a food or production) behind. Approximately because not all improvements are equal. But at the very start, we're probably looking at mines and farms, which are initially +1 improvements.
2. Scout first (rather than second) means puts your Natural Wonder happiness 15-20 points ahead. Scout first greatly increases the chance of finding ruins, which tend to get uncovered by the AI quickly on harder difficulties - ruins are random, but some are exceptionally powerful (for example 30 free culture is enough to immediately take the first Social Policy). Scout first places you 15-20 turns ahead in finding more distant City States - short term that's a huge discovery bonus (25+ gold each), slightly longer term it opens up missions. And finally there's the traditional reason for scouting: Knowledge of terrain and other AIs.
2a. America's +1 visibility arguably makes an early Scout especially powerful, making it much easier to find ruins, natural wonders and other cities. 2 Scouts (before the Worker) may be overkill, but is perhaps worth considering, especially if it looks like your continent is large.
3. Harder difficulties make Barbarians more of a threat early. It will take your city's initial ranged defense 3-4 turns to destroy a Barbarian that's in range. Those are turns when you may not be able to use the best tile, when your Worker will have to immediately retreat to safety, when (if unlucky on positioning) you may even lose an initial improvement to pillaging. On easier difficulties Barbarians will not enter your territory early, so this isn't a consideration. I'm unsure about Emperor, but on demo Deity I've had Barbarians in my city's borders after 20 turns.
4. Given (3), your starting Warrior must remain near your city once your first Worker has been completed. In practice, that makes the Warrior good for a quick circular recon mission, picking up easy ruins and meeting any very close neighbours. So long-distant scouting needs a second unit - a Scout, because Scouts move faster (unless the territory is entirely flat/open), are cheap, and generally surive Barbarians (unless the positioning is really bad).
On easier difficulties, Worker first only makes sense for those that like certainty! The Scout's benefits leave a lot to chance, but the Scout opens many chances for so little trade-off. On harder difficulties the Barbarian threat biases Scout-first even more, because without Warrior protection near your city, a Worker-first advantage will be qucikly lost.
Settlement Location needs to consider Tech Path:
You have water north and east, and mountains north-east, so even if there is land to the north-east, there's a pinch-point there. And you may yet be in a corner. The south-west is clearly exposed, but remember that settling on the current side of the river slows (most) attacks from that direction by a turn. Consider the added protection the river brings (based on what we know of the north/east) before crossing it with your Settler!
It's worth highlighting at this point that positioning isn't as absolutely critical as (say) Civ 4, because of the way borders expand. Consider city location in terms of possibilities and options within 3 hexes of the site, rather than a fixed dot-map.
Silver is a strong opening resource, because they can be worked with 1 tech (Mining). Settling in place would also give Marble quickly, for +10 Happiness. Although I'm not certain how the first tile will be decided in this case - is choice influenced by techs - an auto-pick of Marble first could be frustrating? Sure, you can buy, but that feels like a terrible waste of gold at this stage. I don't have experience of desert floodplains, but the vast amount of desert alone is already pushing that Settler 1 North-North-West, where I see plenty of green and hills, at the cost of (basically) useless desert. The real trade-off is a delay in connecting the Marble, and the risk of a lot of unseen nastiness in the distant north-east fog.
But don't forget tech. Playing America (with late unique units), with a vague (game) objective of playing into the later stages, you're probably going for quite a research and economy-heavy start, rather than a military rush. That probably means Great Library->Civil Service, so you won't want to deviate from that side of the tree much. While you can (and given the Silver, certainly should) delay for Mining (probably even as your very first tech), diverting onto Masonry feels risky.
Take those last 2 paragraphs together, and we might have a strong case for de-prioritising the Marble. Not abandoning it completely, merely accepting that it will be gained later, not early.
Now, the counter-argument here is that since you have Silver and Marble, Mining->Masonry is totally logical. And having made that jump, an early [*cough* I wrote Stonehenge here, but I think I meant] Pyramids is may be too powerful to pass on.
My gut feeling, is not to be distracted by the Marble. Move up (NNE) 1 towards the Silver, so that you are guaranteed that on turn 1, aren't wasting time of defensive advantage by crossing the river. Even unimproved, working that Silver should push your Scout and Worker out sooner, and the sooner those units are done, the sooner the game will really start to take off.
[Added a bit:]
Low Initial Growth:
Above I mentionned the possibility of working a Silver-ed hill immediately, or almost immediately. I should explain the logic:
Based on documentation, unimproved Floodplains+River gives 2F 0H 1G. Unimproved Silver+Hill gives 0F 2H 1G. So the pre-Worker initial trade-off is between food and production. The food slows the first builds, but creates net population. Population adds 1 to research, and here also 1 gold (we have plenty of unimproved floodplains). The production (in this case) stagnates population growth completely, but builds faster.
While immediately stagnating population growth is probably too extreme, there is logic to switching to hill production, and stagnating population growth early - perhaps the second population:
* Food and population has a diminishing return: Each subsequent population growth appears to take longer/more food. Single point research gains contribute a declining percentage increase to research. Population also eats into Happiness, so delays the next Golden Age. In contrast, production is more linear in benefit.
* Production is rare in the early game while food is plentiful. This isn't just due to plentiful food from Maritime City States. Rush-buying production in the early game is (hammer-wise)
more expensive that later. So for early advantage, we should surely focus on what is scarce.
I'm not clear exactly what the optimum strategy is. But it's perhaps worth considering capping population at 2 to get the Worker out 4 or 5 turns quicker, to take maximum advantage of scarce production.
[And some more, since nobody else has posted yet and I'm enjoying the distraction:]
Defenses:
I'll also try and explain why I'm rating defense so highly:
Consider the early mounted units, which get 4 moves, until they hit hills or rivers or trees or anything that isn't open. Now consider your starting terrain:
* The north-east is not only easy to defend with its mountain passes, but full of hills. Nothing is going to take you by surprise there.
* At either end of the NW/SE river appears to be water (coast or lake), so anyone trying to crossings there shouldn't take you by surprise either.
* In contrast you are very vulnerable from the south-west, which has no trees and no blocking hill ranges: An entire army can advance 3 hexes from the fog, and attack, and they'll be nothing you can do to stop it in the early-game.
While the river doesn't protect terribly well against mounted ranged attacks, it pains close-combat units (when actually capturing a city) - either they attack across the river at a penalty, or they cross next to the city and are unable to attack until next turn.
If you're initially playing more of a builder game, the defensive advantages of the capital may be important, depending on who your neighbors are. Whether that's enough to bias the start position away from western suggestions (which look slightly more promising for long-term economy), is unclear.
.
. Thanks a lot on that one Firaxis, I think we found our first thing for the "ugly" bucket 
.
.
.
. I think 1W, buy the Oasis tile, start Mining for our Silver, go Scout->Worker and lay down Farms, then Mines. Its a production capital, I like it.
.
. It pulls in even more hills and gives a pretty clear way to scout for our second city site, I like this move the best so far.
. HOwever, I don't see a Monument as one of the first few builds.
