According to Webopedia:

In Web analytics, including Google Analytics, absolute unique visitor is a "visitor type" report that will count each visitor to your website only once during the date range you have selected. If the visitor was on your site prior to the date range selected this visitor would be sub-classified as a prior visitor. If the visitor did not land on your site prior to date range selected for the report they are classified as a first time visitor.

No, you can't back fill historical web analytics data into Google Analytics, apart from uploading non-AdWords ad cost data. Neither can you remove bad data; it's a bit funny that way.

You might be able to bridge these two sources together by either:

1. Creating a database table or tables that ideally use a set of conformed (consistent) dimension and metric names, i.e. "pageview", "keyword", etc.; SQL can certainly do joins without requiring this, but it'll be a lot nicer to work with if you make sure things are as consistent as possible.
   
   You'll have to decide if you want to use granular data (e.g. individual days, or grouped into months, etc.).
   
   Getting it out of GA is easy using the API and something like this Google Apps Script.

2. Alternatively or in tandem with the above, there are business intelligence (BI) tools that can work with multiple data sources pretty easily, such as Tableau Software and Analytics Canvas, and can map different dimension and metric names to one another. You still want to do your homework before buying anything though, because vendors tend to make things look easier than they really are. AC has a trial and you can is sub. based starting at $49/mo.

Before diving into this however, I would recommend coming up with a handful of important business questions you need answered from the old data, and then working to answer those rather than expecting it to be an entirely seamlessly tied together experience.

Finally, web analytics data's value tends to depreciate over time, once you start collecting good, actionable data, you probably only need a couple of months worth before you can start spending most of your dedicated WA time doing meaningful tracking and analysis.

PS - You may also want to research how well each analytics tool handles false bot/crawler traffic (some can slip into GA), as well as find out what the time of a "visit/session" is; GA begins a new visit/session after 30 minutes of visitor inactivity by default, even if the browser still had the website open. Different tools might treat this differently resulting in different numbers.