I don't know of any case studies, but I remembered reading a couple posts about H1's a couple months back: http://www.seomoz.org/q/multiple-h1-tags-on-same-page

and another from earlier in the year: http://www.seomoz.org/q/one-h1-tag-dead-long-live-multiple-h1-tags

Sorry, I meant to respond to this days ago.  Started, but didn't finish and then it left my mind.

It sounds to me like you want to place a Google map on your site to show the location of the business?  If this is correct, I wouldn't use KML nor KMZ.  I would just use a javascript point market that represents the one location using the Google Maps API.  You can see what I mean as I have several points on one site I'm finishing up at http://www.snoshield.com/contest/christmas2012

You can also show service area like this: http://www.kcdirt.com/contact

Both of those maps use strictly JavaScript, there is no KML or KMZ on them.

I'm not sure if you are referencing SEO in this post, but here's the breakdown from my experience (Several years in the GIS realm)

KMZ when you have a lot of markers in the map making the file size is large and you know the end user will use a Google product to read it.

KML if you are needing to read the file with a program outside of Google due to incompatibility issues with the KMZ format in other programs.

In case you aren't sure what EMD means - Exact Match Domain.

The first thing to remember is that Google will treat the subdomain as a completely different website.  So if you want Google to view your subdomain as it's own site, yes it will bring benefit.  If you are looking for your website to have relevance to the keyword on the domain.com site, then create some content that is relevant to the keyword and put it in a page on the domain - domain.com/page.

Must have been a really good TV show.

There's not much in SEO that is more fun to watch than a competitor take a hit like that. Maybe achieving rank #1 for a competitive term.

Nothing worse in SEO than taking a hit like that.

#3 is the most obvious choice to implement.

All joking aside, I use to run a web design business and the majority of links I had were footer, site wide links.  I still have my site up even though I do very little anymore and haven't focused on any SEO for the site in a long, long time.  After the Penguin/Panda hype, I jumped from page 2 to top 3 consistently for keywords I was once targeting.

I don't know if it was just that my competitors had spammy link profiles or what, but for me in this case site wide footer links seemed to be all I needed.  Take it for what you will.

I personally would go with #1.  Definitely not #4, you never want special characters in the URL.  The reason I say number 1 is because it separates the 1 from the 2 in your 1/2.  #2 could be confused for a 12ct diamond earring, WOW.  #3 I typically avoid underscores in all URLs.

To sum up my choice is #1.  Looks cleanest and when you optimize your page with the 1/2 ct wording, Google is smart enough to see that.  Overall, it won't probably make a huge difference in the end.

1.  I wouldn't remove the ones that have errors and warnings.  I would fix them.  Removing them isn't going to help your other products rank better.  When I mentioned quality of your feed, I just meant don't have a product that is called Blue Mug and then the description says, "Blue Mug"   Title would be Blue Mug, but then describe the mug fully.  This blue mug is 4 inches tall by 2 inches wide and features an ergonomic handle.  It's perfect for the chronic coffee drinker who is afraid of developing carpal tunnel..."  Pretty awesome mug huh?  More descriptive information that is accurate is what I mean by quality.

2.  250 reviews from the 3 different sources from what I can tell is better than all from the same.  The reason I think this is because sometimes it's easy to manipulate one site, but to manipulate 3-4 is more difficult.

3.  From what I've seen Checkout reviews are THE BEST reviews to have, so enabling Google Checkout is important.  They are often times the hardest reviews to get, but the payoff is seen in product listings.

From the testing I have done here are my findings specific to your questions:

1. Age - Not much of a factor.  In fact, I have found that product listings seem to like fresh content.

2. Click Rate - Haven't seen much affect over the 3 sites I run.

3. Overall Quality - BINGO. The quality and quantity of data in your feed is the number 1 thing that I have found truly affects.

4.  Bounce Rate - My bounce rate is pretty low when getting people from product feeds for all my sites, so I can't give feedback if a high bounce rate affects.  I will say that product feed listings is one of the best sources of converting traffic for me though.

5.  Diversity of review sources - This is probably the second biggest factor I have seen combined with #6 Google Checkout reviews.  The more good reviews you have the better your product listings seem to do.

6.  Company location in Google Local - Sorry can't give feedback here.  Location doesn't matter in my niches.

Lastly, I just want to add that you want to make sure that you have data in every field possible.  SKU and other product numbers are very big ones as they help you to rank with the big box boys on products that have several sellers.  Provide something to your customers that they can't, and you'll find it helps a lot.