To generate enough electricity with solar photovoltaics to cover total energy consumption in the USA, you would need land almost equal to the size of Texas.
My brother asked me how much land you would have to cover to satisfy the demand for energy in the USA using renewables. In the previous post I constructed a “back-of-the-envelope” estimate for wind power, finding that it would require a wind farms covering land roughly equal to the size of South Carolina. In this post, I construct a similar analysis for solar power.
The largest U.S. solar farm listed in Wikipedia is Solar Star, but it has not been operational long enough to have good generating statistics posted. I will use the Topaz Solar Farm, which Wikipedia lists as the second largest in the USA.
Topaz is located in San Luis Obispo County, in California’s Central Valley. It is sited on 9.5 square miles, and its average annual generation is 1,100,000 MWh.
https://en.wikipedia.org/wiki/Topaz_Solar_Farm.
The amount of power generated per square mile is 1,100,000 / 9.5 = 115,789 MWh per square mile per year.
In 2014, total energy consumption in the United States was 98,385.2 trillion Btu. = 28,833,750,564 MWh.
http://www.eia.gov/state/seds/data.cfm?incfile=/state/seds/sep_sum/html/rank_use.html&sid=US.
To provide that much power would require 28,833,750,564 / 115,789 = 249,019 square miles.
How to put that in context? It is a square 499 miles on each side. It is just under the size of the state of Texas, it would occupy more than 90% of the state.
https://en.wikipedia.org/wiki/Texas.
Of course, Topaz is located in California’s Central Valley, which has a strong solar resource. The Desert Southwest has an even stronger one, however, and there is a great deal of land there. Still, some of the solar farms would be likely to be spread around the country. That would put some of them in areas with weaker solar resources. In addition, this analysis does not consider the need for excess capacity, redundancy, and storage, all of which would be required to cover times when the sun doesn’t shine, thus requiring even more land. Still, my estimate gives some idea of the size of the task.
Before you boggle at the size of the task, think of our current power generating infrastructure and how long it took us to create it. In 2014 there were an estimated 7,644 power plants in the USA.
https://www.eia.gov/tools/faqs/faq.cfm?id=65&t=2.
The first generating stations supplying power to the public were built in 1882, meaning that it took us 132 years to get to where we are now.
https://en.wikipedia.org/wiki/Power_station.
We have a big job in front of us, but if we give it our best effort, could we, would we, cover that much land with solar panels? I don’t know. But what if you relied on distributed solar photovoltaic power? What if you put solar panels on the roofs of buildings all across America? I will look at that next.
MoGreenStats 
Seeing as you put it this way (and thank you it makes a good illustration) taken literally, The star of Texas would shine awfully brightly. But all that biota would be covered like an Lone-Star parking lot. When I hear of the up and downsides of our energy transition they will always be least- worse compromises. Nuclear, solar farms, wind farms, etc. will ultimately still be blights on landscapes. But of course we haven’t any other option in our haste to arrest climate payback. When I look up at the expanse of the eternal starry sky I regret the parade of communication satellites intruding on the wonder. In your next post I look forward to the diversified grid investigation. It seems a more promising and resilient approach. If only Tesla can fashion an elegant solar roof for our sprawling ticky-tack houses.