Ignoring all the obvious talk about climate "change" itself, this is also almost a perfect example of deceptive framing.

Because 2014 is literally the time Twitter started becoming a semi-major site to begin with. Prior to that, it had barely 100k tweets per day total across every single topic. So using a pure numerical Y axis is a complete piece of propaganda because it loses that context.

Not only that, but its not based on the discussion in the broadest terms possible to measure the conversation. Only on specific terms they decided are strongly associated with a specific strawman of their opponents, which given how language evolves those terms probably weren't even in common usage prior.

They even openly admit to excluding a specific event for some reason, probably because it would either show the number dropping recently from some 2021 peak or even just having a natural up curve from 2019 until now.

This isn't even getting into the idea of bots, shills, and obvious plants. Something that we know has been a thing forever and would only come into place once the opposition started the conversation in the first place.