I'd almost worry more about about the size of the panels making installation problematic in a light to moderate wind, and having them sail off the roof into the neighbor's car.
Nowhere close to mining coal of course.
http://www.forbes.com/sites/jamesconca/2012/06/10/energys-de...
Does anyone more knowledgeable care to comment? What other estimates are there? How are they calculated?
Additionally, I'd like to see fatality numbers for solar overall, not just rooftop solar. In the US at least more solar capacity is installed at ground-mount plants than on rooftops. The rooftop number itself could probably use an update; increasing module efficiency, if nothing else, is going to reduce the number of rooftop trips required for a given capacity, hence the number of accidents per terawatt hour.
That said, the combination of the PowerWall and the roof might actually make it more amenable for power companies since you could grid tie "for emergencies" but lower your total overall power budget and help avoid the evening surge (max grid usage in the Bay Area is about 5:30PM, solar is off-line and people are coming home from work, firing up appliances (like cooking dinner, getting a beer from the fridge which turns on the converter) environment systems (heating/cooling) are activating to bring the house to the 'occupied' temperature, Etc.
If your house uses its PowerWall for all that energy you'll appear as a much smoother non-load on the grid.