I can see your problem when i do the crawl, when i look at the page I can not see the offending links in the code.

So we can say it happaning dynamicaly. Try disabling addons. I have seen simullar urls in wordpress sites i have crawled.

The question is why do it?

Some say that the H1 has little value, but thats not the case, goolge will try to find a heading and i believe SEOMoz test show that using a H1 had lttle benifit over other ways of presenting a header, but all the same a header you should have and the best header is the H1.

Bing on the over hand dose value the H1 tag and sees no H1 tag as a violation. Bing says the H1 is almost as important as the Title tag.

http://perthseocompany.com.au/seo/reports/violation/the-h1-tag-is-missing

It also stats that muliplue use may lead to no indexing, I would not mess with it, expecialy if there is no reason to do so.

http://perthseocompany.com.au/seo/reports/violation/the-page-contains-multiple-h1-tags

I cant see a redirection doing this, i would put the redirection back, it is more likely some other reason.

i could not find any rel=next or prev in those pages, there was a rel canonical.

If you do a rel canonuical then you need to have the content on the canonical page. You can not have a page with ABC the canonical it to a page with XYZ.

If you use rel next and prev. the fiorst page will have rel=next the last will have rel=prev, all pages in between will have both. Google will give credit to the first page with only rel=next, all links pointing to any of the pages will pass link juice to that first page.

http://googlewebmastercentral.blogspot.com/2011/09/view-all-in-search-results.html

http://googlewebmastercentral.blogspot.com/2011/09/view-all-in-search-results.html

I remember seeing a video where Matt Cutts said they were useless, I think he mentioned articles posting, blog comments and directories. He said that local and niche Directories mixed with a few quaility links still had value, the others were of no value.

honestly, I think you are wasting your time and may even be doing damage.

More then likely they are no follow, even if they are not they are very low quality links, more likely no value at all.

I would not waste my time.

If you can no-index , follow all but the default, then you will send link juice to the pages but it will return the link juice because it is follow, but they will not index because they are no-index.

If you use robots, then it can not read the page to follow the links.

You can, but then you have links passing link juice to non followed pages. it would be better if you used canonical. even better would be to add no-index, follow meta tag when non canonical page is displayed, but this requres some codeing.