The problem with looking for scientific support is it comes in tiny pieces. Going back 30 years you find the early stuff is looking at the general trends but people don't republish the same report looking at the same data. So if you want to get a solid understanding of what scientists think you really need to read a few hundred technical reports each of which is a tiny slice of the pie. You also need to verify that each of those reports is based on solid science and created by people without significant bias; luckily if you go far enough back the issue is not all that controversial so the age of the publication is a good indicator of quality.

It would be nice to have a reliable way to avoid that much work, but if you try to shortcut this you are going to be faced with summery information that's of questionable quality.