Show me any form of serious progress on the energy storage problem, please. Tesla's $40,000 car battery is about the best there is, and that's not encouraging.
In addition for evening out dips and bumps in unreliable renewable energy, it's a good option for excess using excess capacity in nuclear power plants in order to deal with load variation. (Nuclear reactors don't like changing their energy production levels, so evening out the load on them by buffering with pumped water energy storage is a great benefit).
Flywheels are also another option, although I don't think they're nearly as popular.
Overall, purely mechanical solutions seem to be preferred over chemical energy storage for large installations.
Flywheel storage is OK for spinning reserve (responding to fluctuations on the order of an hour or two) but is too expensive otherwise.
My money is on thermal storage, and cheap-substrate batteries (molten salt, liquid metal) for longer-scale storage. For national scale grid storage there are many alternatives for which there's simply not sufficient material to create a feasible alternative.
I haven't really heard of solar based "atmospheric disruption" issue. On the other hand, I have wondered what the effect of releasing gigawatts (terawatts?) of heat (and electricity that will become heat) derived from nuclear power plants year after year. Solar is a closed system energy source; we're going to receive that energy one way or another whereas nuclear is adding energy to the system derived by converting matter - energy that isn't a normal input.
If cities at large become net exporters of electricity it would in some small part alleviate heat bubbles.
If the panels are next the the city, you do concentrate heat more in the city as electric is delivered, but no more than electric supplied and consumed from non solar sources. If the panels are adjacent to the city, you again alleviate the heat bubble to some small extent because you're cooling one area and increasing the thermal gradient between the hot city and cooler solar panel farm.
I don't know how one would conclude that any of these effects are of a magnitude to "wreck the local ecosystem".
http://en.wikipedia.org/wiki/Grid_energy_storage
The other solution is large-scale intelligent power grids. The wind always blow somewhere and the sun always shines somewhere. The amount of grid-scale storage backing required for uninterrupted power is probably less than 12 hours. I'm sure someone has done a solid analysis on this.
Sure, but then your power supply is in the hands of other countries. What happens during war?
The question is: is it better to spend billions researching new generations of nuclear reactors or to spend billions working on energy storage?
The former will lead to a switch from older, less concentrated fossil fuels to a newer, more energy-dense fossil fuel. But it's not without well-known problems.
The latter will result in a permanent, eternal solution to the energy problem.
There are limits to how much solar and wind is available, but those are orders of magnitude beyond our civilization's current energy requirements. Other limits to growth -- arable land, decreasing fertility with increasing wealth -- will likely prevent us from ever even approaching those limits.
Solve the energy storage problem and the industrial revolution becomes as permanent as the agricultural revolution. Nuclear power won't do that. So energy storage technology is IMHO a more prudent, higher dividend investment than better nuclear reactors.