Figure 1.
Variation of average sunshine with latitude and with time of year. (a) Average power of sunshine falling on a horizontal surface in selected locations in Europe, North America and Africa. These
averages are whole-year averages over day and night. (b) Average solar intensity in London and Edinburgh as a function of time of year. (Average powers per unit area are sometimes
measured in other units, for example kWh per year per square metre; for the reader who prefers those units, the following
equivalence may be useful: 1 W = 8.766 kWh per year.) Sources: NASA's Surface meteorology and Solar Energy (eosweb.larc.nasa.gov;
www.africanenergy.com/files/File/Tools/AfricaInsolationTable.pdf; www.solarpanelsplus.com/; solar-insolation-levels/lightbucket.wordpress.com/2008/02/24/insolation-and-a-solar-panels-true-power-output/.)
Figure 2.
Electricity, gas and transport demand; and modelled wind production, assuming 33 GW of capacity, all on the same vertical
scale. Wind production is modelled by scaling data from Ireland.
Figure 3.
Power consumption per person versus population density, in 2005. Point size is proportional to land area (except for areas
less than 38 000 km2 (e.g. Belgium), which are shown by a fixed smallest point size to ensure visibility). The straight lines with slope -1 are
contours of equal power consumption per unit area. Seventy-eight per cent of the world's population live in countries that
have a power consumption per unit area greater than 0.1 W m-2. (Average powers per unit area are sometimes measured in other units, for example kWh per year per square metre; for the
reader who prefers those units, the following equivalence may be useful: 1 W= 8.766 kWh per year.)
[See also
David MacKay's Map of the World]
Figure 4.
Power consumption per person versus population density, in 2005. Point size is proportional to land area. Line segments show
15 years of 'progress' (from 1990 to 2005) for Australia, Libya, the USA, Sudan, Brazil, Portugal, China, India, Bangladesh,
the UK and the Republic of Korea. Seventy-eight per cent of the world's population live in countries that have a power consumption
per unit area greater than 0.1 W m-2.
[See also
David MacKay's Map of the World]
Figure 5.
Power consumption per person versus population density, from 1600 or 1800 to 2005. OECD, Organization for Economic Cooperation
and Development.
Figure 6.
Power consumption per person versus population density, in 2005. Point size is proportional to land area. The diagonal lines
are contours of power consumption per unit area. The grey box corresponds to the region shown in figures 3 and 4.
[See also
David MacKay's Map of the World]
Figure 7.
Electricity production from AllEarth Renewables Solar Farm, 350 Dubois Drive, South Burlington, VT (latitude 44°26' N), during the last six months of 2011 and the first six months of 2012; and insolation (10 year average) for Montpelier
(33 miles away from the farm) from the NASA Surface meteorology and Solar Energy Data Set. Photo courtesy of AllEarth Renewables.
Figure 8.
Solar farms' average power per unit land area versus the local insolation (i.e. average incident solar flux per unit of horizontal
land area). Filled triangles, squares, circles and pentagons show ground-based solar photovoltaic farms. The other point styles
indicate roof-mounted photovoltaic farms and solar thermal facilities. Where the solar farm name is shown in black, actual
electricity-production data have been displayed; otherwise, for names in grey, the electricity production is a predicted value.
(See tables 1-3 for data.) Both axes show average power per unit area, averaging over the whole year including day and night. (Average powers
per unit area are sometimes measured in other units, for example kWh per year per square metre; for the reader who prefers
those units, the following equivalence may be useful: 1 W=8.766 kWh per year.)
Figure 9.
Solar farms' load factors versus their insolation. The grey lines show, as guides to the eye, the relationships
(load factor)/(insolation/(1000 W m-2)) = {1.33, 1.0, 0.67}.
Figure 10.
Solar farms' average power per unit land area versus their load factor (i.e. the ratio of their average electrical output
to their capacity). Three of the Spanish thermal solar electric power stations have load factors greater than 27% and therefore
fall off this chart to the right. (See tables 1-3 for data.)
Figure 11.
Electricity demand in the UK and modelled solar production, assuming 40 GW of solar capacity. (a-c) The upper curves show Britain's electricity demand, half-hourly, in 2006. The lower data sequence in (a) is a scaled-up rendering of the electricity production of a roof-mounted south-facing 4.3 kW 25 m2 array in Cambridgeshire, UK, in 2006. Its average output, year-round, was 12 kWh per day (0.5 kW). The data have been scaled
up to represent, approximately, the output of 40 GW of solar capacity in the UK. The average output, year round, is 4.6 GW.
The area of panels would be about 3.8 m2 per person, assuming a population of roughly 60 million. (For comparison, the land area occupied by buildings is 48 m2 per person.) (b,c) The lower curves show, for a summer week and a winter week, the computed output of a national fleet of 40 GW of solar panels,
assuming those panels are unshaded and are pitched in equal quantities in each of the following 10 orientations: south-facing
roofs with pitch of (1) 0°, (2) 30°, (3) 45°, (4) 52°, and (5) 60°; (6) south-facing wall; and roofs with a pitch of 45° facing (7) southeast, (8) southwest, (9) east and (10) west. On each day, the theoretical clear-sky output of the panels
is scaled by a factor of either 1, 0.547, or 0.1, to illustrate sunny, partially sunny, and overcast days. Note that, on a
sunny weekend in summer, the instantaneous output near midday comes close to matching the total electricity demand. Thus,
if solar photovoltaics is to contribute on average more than 11% of British electricity demand without generation being frequently
constrained off, significant developments will be required in demand-side response, large-scale storage, and interconnection.
Figure 12.
Contour plot of the total cost of a photovoltaic system, in a sunny location, capable of giving a steady 1 kW output with
(a) 14 h of storage (as might be appropriate in a location such as Los Angeles); (b) 120 h of storage (as might be appropriate in cloudier locations), as a function of the cost of the panels and the cost of
storage. Assumptions: load factor, 20%; efficiency of electrical storage, 75%; fraction of final electricity that comes through
the store, 60%. The capital costs per kW are equivalent to the following undiscounted costs per kWh, assuming 20 years' operation:
Costs of battery storage are from Poonpun & Jewell. Cost of pumped storage (p.s., $125 per kWh) is based on Auer & Keil. The cost of the Vermont solar farm (section 3), built in 2011, was $5630 per kW of capacity ($12 million for 2130 kW),
without electricity storage. Note that the total cost of this solar farm is more than three times the cost of its photovoltaic
modules (roughly $1750 per kW).
Figure 13.
Contour plot of potential average consumption of electrical power as a function of production and energy intensity of storable
materials. The points show these two properties for six materials: ice, ammonia, aluminium, hot water, hydrogen and gasoline
from thin air. Where there are two points, the right-hand coordinate indicates proven achievable energy intensity of production,
and the left-hand coordinate shows the conceivable energy intensity with efficiency improvements. For ice, ammonia, aluminium,
hot water and hydrogen, the production shown is today's production; the arrows indicate levels to which production could rise
if stored ice were used as a carrier of cold for air-conditioning, if stored water were used as a carrier of heat for space-heating,
and if hydrogen took a significant role in transport. For gasoline production from air, the 'production' shown is today's
per capita consumption of transport fuels in the UK.
Figure 14.
Solar farms' average power per unit land-area versus their capacity. (See tables 1-3 for data.)