Possibility to select the Site content itself for Canonical content

[Bellfalasch]

We use ?utm_source=x and similar parameter URLs to track ads, and more. However we have had multiple reports from SEO-firms that these are picked up by Google and other webcrawlers as potential multiple pages. So, a solution for us is to actively use the canonical meta-data to make sure spiders only track the absolutely correct URL (no trailing slash, no URL parameters, etc). This works for all sub-pages, but not for the main site content since it has a <allowPath>${site}/*</allowPath> making the site itself unselectable in the list. Is this fixable?