Exploring the Development of Solar Energy in California

Solar PV generation has become an integral part of the renewable energy industry. With state-level, renewable portfolio standards in place, solar power demand has substantially increased and become a competitive and economically viable energy solution throughout the world. California has one of the strongest renewable goals of 50% renewable generation by 2030. This is encouraging the increase adoption of solar power as a form of renewable energy. Especially, with the decline in price for PV equipment, solar power has become more affordable. However, this was not always the case with solar. Therefore, this research explores the connection that average household income has on the megawatts of solar arrays installed by county. By obtaining data from Go Solar California, a table of all installed solar installations by megawatt capacity in California, was imported into Carto (all_solar_installations_by_county). This dataset was used to join with the cb_2013_us_county_500k dataset to assign “the_geom” to each county and the megawatts of solar installed by using this query:

SELECT
counties.cartodb_id,
counties.the_geom_webmercator,
solar.megawatts,
solar.county,
solar.countyfp,
solar.number_of_applications
FROM
cb_2013_us_county_500k as counties
INNER JOIN
all_solar_installations_by_county as solar
ON counties.countyfp=solar.countyfp
WHERE
counties.statefp = '06'

This query created the dataset: mw_solar_by_county that was then used to find which counties in this list of top 20 solar installations also had the greatest average household income. Poverty census data that also contained the average household income was imported to Carto as poverty_ca_county_the_geom and this query was used to find where these counties overlapped:

SELECT
solar.cartodb_id,
solar.the_geom_webmercator,
solar.megawatts,
solar.county,
solar.countyfp,
solar.number_of_applications,
poverty.all_ages_in_poverty_count,
poverty.median_household_income_in_dollars
FROM
poverty_ca_county_the_geom as poverty,
mw_solar_by_county as solar
WHERE
St_Contains(
poverty.the_geom_webmercator,
solar.the_geom_webmercator)

As you switch between layers in this map you can see that there is not a connection between the average household income and megawatts of solar capacity for each county:

Since this did not show a direct connection the megawatt capacity of residential solar installations was compared to average household income instead. The original dataset included both residential and commercial installations. The resi_solar_installations dataset was also obtained from Go Solar California and joined again to the counties dataset:

SELECT
counties.cartodb_id,
counties.the_geom_webmercator,
resi_solar.megawatts,
resi_solar.county,
resi_solar.countyfp,
resi_solar.number_of_applications
FROM
cb_2013_us_county_500k as counties
INNER JOIN
resi_solar_installations as resi_solar
ON counties.countyfp=resi_solar.countyfp
WHERE
counties.statefp = '06'

This dataset was created, resi_solar_installations_county, and the same query was used to connect the average household income dataset to the residential solar installations dataset:

SELECT
resi_solar.cartodb_id,
resi_solar.the_geom_webmercator,
resi_solar.megawatts,
resi_solar.county,
resi_solar.countyfp,
resi_solar.number_of_applications,
poverty.all_ages_in_poverty_count,
poverty.median_household_income_in_dollars
FROM
poverty_ca_county_the_geom as poverty,
resi_solar_installations_county as resi_solar
WHERE
St_Contains(
poverty.the_geom_webmercator,
resi_solar.the_geom_webmercator)

This map shows that there is also not a connection between only residential installations and average household income:

This is likely due to the fact that solar has matured as an industry and become economically competitive. Therefore, solar has become more widespread among all communities and not only families with the highest household income.

Next we will take a look at the solar radiation in areas with the highest capacity of solar installations. Annual net surface radiation data was obtained from Cal Adapt and imported as a table into Carto as this dataset: ca_annual_solar_radiation_1. The x and y coordinates in this dataset were used to create a geom column that could be connected to the a megawatts of solar capacity table using this query:

SELECT
cartodb_id, solar_radiation, data_source, model_id, scenario_id, point_id, x, y,
ST_SetSRID(    ST_MakePoint(
x,
y
),
4326
) AS the_geom
FROM (SELECT * FROM ca_annual_solar_radiation_1) AS _camshaft_georeference_long_lat_analysis

The ca_annual_solar_radiation_the_geom dataset was created from this query and used to perform an ST_Intersction:

SELECT
ca_annual_solar_radiation_the_geom.*,
mw_solar_by_county.megawatts,
mw_solar_by_county.county
FROM
ca_annual_solar_radiation_the_geom,
mw_solar_by_county
WHERE
ST_Intersects(
ca_annual_solar_radiation_the_geom.the_geom_webmercator,
mw_solar_by_county.the_geom_webmercator
)
ORDER BY solar_radiation DESC

This map was created to show the relevance of solar radiation potential on increased solar capacity throughout California

In general, there does seem to be a direct connection between the available solar radiation and the megawatts of solar projects installed, along with the area of that county. For example, San Bernardino appears to receive a high level of annual net surface radiation along with and already established solar market that has a high capacity of solar installations.

Other factors that could be affecting this growth in the California Solar market include policy development that is no longer provides retail rate incentive credits for feed-back into the grid and is encouraging the adoption of solar energy storage instead of increased solar capacity.