In 2007, when the cost of a rooftopsolarsystem hovered around $8 per watt, engineer and solar advocate Bill Powers proposed an ambitious plan: replace a proposed transmission line for Southern California with 2,000 megawatts of PV systems across San Diego.

Opponents of the Sunrise Powerlink transmission line, which was completed by the California utility SDG&E in 2012, seized on the report as proof that distributed solar could offset the need for large-scale infrastructure projects built in sensitive habitats. Powers' report claimed that commercial rooftop solar systems could supply the same amount of energy as the transmission project, for roughly the same cost.

Even though California had already established a plan to build 3,000 megawatts of rooftop PV across the state, the renewable energy industry at that time was focused on achieving scale. That meant more attention on "transmission superhighways" that could connect mega-scale projects in rural areas to urban load centers.

The Sunrise Powerlink line itself was pitched to regulators as a way to transmit electricity from big geothermal, solar and wind projects in the Imperial Valley to San Diego. SDG&E says it has connected 1,000 megawatts of renewable energy capacity to the line since it was constructed.

Today, however, solar PV is in a much different place than it was eight years ago. According to the latest data from GTM Research, average residential PV system prices hit $3.48 per watt in 2014. The residential sector is now growing faster than any other PV sector.

In the commercial space, average system prices hit $2.25 per watt by the end of last year. Utility-scale solar is still the cheapest option, with prices ranging from $1.40 to $2.10 per watt. 

With prices and installation costs for all system types continuing their downward trend, some are once again asking the question: does it make more sense to rely on distributed projects on buildings for our energy needs?

A new study published in the journal Nature Climate Change attempts to provide an answer for the entire state of California. Written by Stanford University researchers, the study concludes that solar PV in "compatible" areas within the built environment could provide the state with 4.8 to 6.5 times more electricity generation than the state consumed in 2011.

"Comparing the realized, generation-based potential of solar energy technologies to the state of California’s total energy consumption further underscores the value of solar. The quantity of energy that could be produced solely within the built environment (that is, ‘compatible’; conferring the least land-use or land-cover change) exceeds the energy needed to meet the state’s total energy consumption," write Rebecca Hernandez, Madison Hoffacker and Christopher Field.

They also conclude that dry-cooled concentrating solar power in developed areas close to transmission lines (urban open space) could meet California's electricity needs 2.7 times over.

Unlike Bill Powers' 2007 study, this new analysis doesn't look at the cost of such a plan. Instead, the researchers modeled the technical potential of solar and available space in the built environment, and then categorized how suitable the different areas are for PV and CSP projects.

According to the study, rooftop and ground-mounted PV systems on compatible spaces could generate between 10,600 and 14,600 terawatt-hours per year, and CSP systems could generate 5,947 terawatt-hours per year. Areas designated as "potentially compatible" offer threefold more generation potential.

Of course, technical potential doesn't say much about system-wide compatibility and flexibility -- and in turn, the total cost.

Pointing to the so-called "duck curve," California's grid operator is already warning about solar overgeneration in the state by 2020. As more intermittent renewables come on-line, the California Independent System Operator (CAISO) says that utilities will need to ramp 13,000 megawatts of power plants in order to make up for declining solar and wind generation during the late-day peak. As Julie Blunden recently pointed out, the "duck" is maturing five years ahead of CAISO's projections.

There are many ways to address this ramping issue. Thermal storage, long- and short-duration battery storage, demand charges and demand response programs are all potentially competitive replacements for conventional power plants. But the cost of implementing these technologies and balancing the grid are just as crucial as understanding the technical potential of solar itself.

The researchers aren't attempting to address that integration challenge. They simply show that California's existing built environment is more than enough to meet the state's yearly electricity needs with solar.

And, in so concluding, they believe their study addresses another challenge restricting renewables development: land limitations.

"Solar energy within the built environment may be an overlooked opportunity for meeting sustainable energy needs in places with land and environmental constraints," write the researchers. "These areas provide options for minimizing environmental impacts associated with a large-scale transition to a renewable energy mix."