Bitcoin mining is uniquely placed to facilitate and enhance the move towards 100% renewable energy.
Another consideration regarding this point:
If you want to power something (home, business, etc.) 100% from solar, you need to hugely oversize the installation to harvest enough energy on overcast winter days.
However, this means in summer you are going to have tons of excess energy, making it financially infeasible.
There are already services that buy your excess energy (even though usually way underpriced), but in a future where most people have solar, there will probably be so much excess energy at summer noon that there will be no buyer.
Either mining yourself with that energy that nobody wants, or selling it to other people who are mining, is a great idea.
This is an often brought up argument, but one counter-argument that I can think of is that when done at 'small scale' you will run into the issue of how to power the ASICs when you don't have this excess power. Turning them off is technologically challenging (faster hardware degradation) and insanely elongates ROI, so it's not really an option either with current-gen machines.
I guess that in those times where you don't have excess energy (mostly in winter) you can actually use the heat. Maybe, there will be ASICs tailored to being frequently turned on and off.