Climate Change 2001:
Working Group III: Mitigation
Other reports in this collection

3.8.4.3.4 Solar Energy

An estimation of solar energy potential based on available land in various regions (Tables 3.33a and 3.33b) gives 1,575 to 49,837 EJ/yr. Even the lowest estimate exceeds current global energy use by a factor of four. The amount of solar radiation intercepted by the earth may be high but the market potential for capture is low because of:

  1. the current relative high costs;
  2. time variation from daily and seasonal fluctuations, and hence the need for energy storage, the maximum solar flux at the surface is about 1 kW/m2 whereas the annual average for a given point is only 0.2 kW/m2;
  3. geographical variation, i.e. areas near the equator receive approximately twice the annual solar radiation than at 60° latitudes; and
  4. diffuse character with low power such that large-scale generation from direct solar energy can require significant amounts of equipment and land even with solar concentrating techniques.
Table 3.33a: Key assumptions for the assessment of the solar energy potential
Region
Assumed annual
clear sky irradiancea
kW/m2
Assumed annual
average sky
clearanceb, %
 
Min
Max
Min
Max
NAM (North America)
0.22
0.45
0.44
0.88
LAM (Latin America and the Caribbean)
0.29
0.46
0.48
0.91
AFR (Sub-Saharan Africa)
0.31
0.48
0.55
0.91
MEA (Middle East and North Africa)
0.29
0.47
0.55
0.91
WEU (Western Europe)
0.21
0.42
0.44
0.80
EEU (Central and Eastern Europe)
0.23
0.43
0.44
0.80
FSU (Newly independent states of the former Soviet Union)
0.18
0.43
0.44
0.80
PAO (Pacific OECD)
0.28
0.46
0.48
0.91
PAS (Other Pacific Asia)
0.32
0.48
0.55
0.89
CPA (Centrally planned Asia and China)
0.26
0.45
0.44
0.91
SAS (South Asia)
0.27
0.45
0.44
0.91
a The minimum assumes horizontal collector plane; the maximum assumes two-axis tracking collector plane
b The maxima and minima are as found for the relevant latitudes in Table 2.2 of WEC (1994).

Table 3.33b: Assessment of the annual solar energy potential
Region

Unused land

(Gha)

Assumed for
solar energyd
(Mha)

Solar energy
potentiale
(EJ/yr)
 
Availablec
Min
Max
Min
Max
NAM (North America)
0.5940
5.94
59.4
181.1
7,410
LAM (Latin America and the Caribbean)
0.2567
2.57
25.7
112.6
3,385
AFR (Sub-Saharan Africa)
0.6925
6.93
69.3
371.9
9,528
MEA (Middle East and North Africa)
0.8209
8.21
82.1
412.4
11,060
WEU (Western Europe)
0.0864
0.86
8.6
25.1
914
EEU (Central and Eastern Europe)
0.0142
0.14
1.4
4.5
154
FSU (Newly independent states of the former Soviet Union)
0.7987
7.99
79.9
199.3
8,655
PAO (Pacific OECD)
0.1716
1.72
17.2
72.6
2,263
PAS (Other Pacific Asia)
0.0739
0.74
7.4
41.0
994
CPA (Centrally planned Asia and China)
0.3206
3.21
32.1
115.5
4,135
SAS (South Asia)
0.1038
1.04
10.4
38.8
1,339
World total
3.9331
39.33
39.33
1575.0
49,837
Ratio to the current primary energy consumption (425 EJ/yrf)
-
-
-
3.7
117
Ratio to the primary energy consumption projectedg for 2050 (590-1,050 EJ/yr)
-
-
-
2.7 - 1.5
84 - 47
Ratio to the primary energy consumption projectedg for 2100 (880-1,900 EJ/yr)
-
-
-
1.8 - 0.8
57 - 26
c The "other land" category from FAO (1999)
d The maximum corresponds to 10% of the unused land; the minimum corresponds to 1% of the unused land
e The minimum is calculated as (9) = (2)x(4)x(7)x 315 EJ/a, where numbers in parentheses are column numbers in Tables 3.33a and 3.33b, 315 is a coefficient of unit conversion; the maximum (10) is (3)x(5)x(8)x315 EJ/yr
f Source: IEA (1998b)
g Source: IIASA/WEC (1998)

Photovoltaics

The costs of photovoltaics are slowly falling from around US$5,000/kW installed as more capacity is installed in line with the classical learning curve (Goldemberg, 2000). Present generating costs are relatively high (20 – 40c/kWh), but solar power is proving competitive in niche markets, and has the potential to make substantially higher contributions in the future as costs fall. Photovoltaics can often be deployed at the point of electricity use, such as buildings, and this can give a competitive advantage over power from central power stations to offset higher costs.

Conversion technology continues to improve but efficiencies are still low. Growing markets for PV power generation systems include grid connected urban building integrated systems; off-grid applications for rural locations and developing countries where 2 billion people still have no electricity; and for independent and utility-owned grid-connected power stations. The size of the annual world market has risen from 60MW in 1994 to 130MW in 1997 with anticipated growth to over 1000MW by 2005 (Varadi, 1998). This remains small compared with hydro, wind, and biomass markets. Industrial investment in PV has increased with Shell and BP-Solarex establishing new PV manufacturing facilities with reductions in the manufacturing costs anticipated (AGO, 1999).

Conversion efficiencies of silicon cells continue to improve with 24.4% efficiency obtained in the laboratory for monocrystalline cells and 19.8% for multicrystalline (Green, 1998; Zhao et al., 1998), though commercial monocrystalline-based modules are obtaining only 13%-17% efficiency and multicrystalline 12%-14%. Modules currently retail for around US$4,000 – 5,000/kW peak with costs reducing as predicted by the Worldwatch Institute (1998) as a result of manufacturing scale-up and mass production techniques. Recent studies showed a US$660M investment in a single factory producing 400MW (5 million panels) a year would reduce manufacturing costs by 75%. KPMG (1999) and Neij (1997) calculated a US$100 billion investment would be needed to reach an acceptable generating level of US$0.05/kWh.

Thin film technologies are less efficient (6%-8%) but cheaper to produce, and can be incorporated into a range of applications including roof tile structures. Further efficiency improvements are proving difficult, whereas both cadmium telluride and copper indium gallium selenide cells have given 16%-18% efficiencies in the laboratory (Green et al., 1999) and are close to commercial production. New silicon thin film technology using multilayer cells, which combine buried contact technology with new silicon deposition and recrystallization techniques, enables manufacture to be automated. A commercially viable product now appears to be feasible with an efficiency of around 15% and cost of around US$1500/kW (Green, 1998). Recycling of PV modules is being developed at the pilot scale for both thin film and crystalline silicon modules (Fthenakis et al., 1999).

Advances in inverters (including incorporation into the modules to give AC output) and net metering systems have encouraged marketing of PV panels for grid-connected building integration projects either in government sponsored large scale installations (up to 1MW) or on residential buildings (up to 5kW) (IEA, 1998c; Moomaw et al., 1999b; IEA, 1999b; Schoen et al., 1997). Japan aims to install 400 MW on 70,000 houses by 2000 (Flavin and Dunn, 1997) and 5000MW by 2010. Simple solar home systems with battery storage and designed for use in developing countries are being installed and evaluated in South Africa and elsewhere by Shell International Renewables with funding from the World Bank. Integrated building systems and passive solar design is covered in Section 3.3.4.

A promising low-cost photovoltaic technology is the photo-sensitization of wide-band-gap semiconductors (Burnside et al., 1998). New photosensitizing molecules have been developed in the laboratory, which exhibit an increased spectral response, though at low efficiencies of <1%. Arrays of large synthetic porphyrin molecules, with similar properties to chlorophyll, are being developed for this application (Burrell et al., 1999).

Solar Thermal

In Europe 1 million m2 of flat plate solar collectors were installed in 1997, anticipated to rise to 5 million m2 by 2005 (ESD, 1996). Combined PV/solar thermal collectors are under development with an anticipated saving in system costs, though these remain high at US$0.18-0.20/kWh at 8% discount rate and 10 year life (Elazari, 1998). High temperature solar thermal power generation systems are being developed to further evaluate technological improvements (Jesch, 1998). The Californian “power tower” pilot project has been successful at the 10 MW scale and is now due to be tested at 30MW with 100MW the ultimate goal (EPRI/DOE, 1997). Dish systems giving concentration ratios up to 2000 and therefore performing at temperatures up to 1,500oC can supply steam directly to a standard turbo-generator (AGO, 1998). Capital costs are projected to fall from US$4,000/kW to US$2,500 by 2030 (Moomaw et al., 1999b) with other estimates much lower (AGO, 1998).

3.8.4.3.5 Geothermal

Geothermal energy is a heat resource used for electricity generation, district heating schemes, processing plants, domestic heat pumps, and greenhouse space heating, but is only “renewable” where the rate of depletion does not exceed the heat replenishment.

The geothermal capacity installed in 20 countries was 7,873 MWe in 1998: this provided 0.3% (40TWh/yr) of the total world power generation (Barbier, 1999). Geothermal direct heat use was an additional 8,700 MWth. This energy resource could be increased by a factor of 10 in the near term with much of the resource being in developing countries such as Indonesia (Nakicenovic et al., 1998).

3.8.4.3.6 Marine Energy

The potential for wave, ocean currents, ocean thermal conversion, and tidal is difficult to quantify but a significant resource exists. For example, resources of ocean currents greater than 2 m/s have been identified, and in Europe alone the best sites could supply 48TWh/yr (JOULE, 1993). Technical developments continue but several proposed schemes have met with economic and environmental barriers. Many prototype systems have been evaluated (Duckers, 1998) but none have yet proved to be commercially viable (Thorpe, 1998).

Several ocean current prototypes of 5 to 50kW capacity have been evaluated with estimated generating costs of around US$0.06-0.11/kWh (5% discount rate) depending on current speed, though these costs are difficult to predict accurately (EECA, 1996). The economics of tidal power schemes remain non-viable, and there have been environmental concerns raised over protecting wetlands and wading birds on tidal mudflats.



Other reports in this collection