Energy Choices in a Democratic Society: The Report of the Consumption, Location, and Occupational Patterns Resource Group, Synthesis Panel of the Committee on Nuclear and Alternative Energy Systems, National Research Council. (1980)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 TO SUPPLY ENERGY: THE SUN Chapter 3's discussion of the breeder reactor drew upon the themes of safety, liberty, and institutional structure because the decisions about the nuclear breeder will no longer be determined solely on the scien- tific engineering and economic grounds that have in the past been central. This chapter's discussion of solar technologies includes consideration of feasibility issues, costs, organizational constraints, and societal impacts of solar energy because these are the variables that are dis- cussed increasingly in any debate on what is to happen after fossil fuels can no longer serve as our main energy sources. Until recently, solar energy and renewable technologies did not receive serious attention. In spite of the obvious fact that the sun shines every day, considerations of solar energy have been eclipsed by the focus on more conventional, high technologies such as nuclear fis- sion. The State of Connecticut, though, receives approximately as much energy from the sun each year as the entire nation used from all nonfood forms of energy during 1972. To phrase it differently, and more force- fully, one barrel of oil contains as much chemical energy as the sun supplies to each square meter of land in the United States each year. There is no question that solar energy is abundant now and that it will be available to human beings for as long as we might need it. In spite of this, no one can say with assurance what energy impact solar technology would have on the U.S. economy. The writers of this report believe that if a national decision were made to encourage a rapid transition to renewable resources, solar energy could be far more significant than most energy forecasts indicate. As we view the matter, most of the forecasts to date were made by groups with a strong technol- ogy orientation; they have emphasized high-technology uses for solar energy, mainly for generating electricity, which we believe to be the least economic use of solar energy at present. 37

38 On the contrary, we emphasize other sorts of uses for solar energy and compare our view with those of other forecasters.3 Unfortunately, such forecasters rarely speak about the social impact of solar-energy technology or its consequent effect on social organization. Although the cost of energy development involves more than Btu's and dollars per Btu, we have developed our discussion within conventional boundaries— cost and use—in order to set the stage for reader participation in the scenario chapters that follow. The assumption that undergirds the sce- nario chapters, however, is that society is interconnected and evolves systemically. A shift in the direction of husbanding energy resources through energy conservation, an alteration in consumption habits, the development of renewable energy sources such as solar and geothermal— all these need to be seen in a systemic context. Energy systems impinge upon economic growth, the prospects of war and peace, and the degree of tension between countries. But in this chapter on solar power, we turn to some rather straightforward questions: To what uses can solar energy be put? How soon can solar technology be available? How much would solar systems cost? Are there geographic variations? What are the im- plications of decentralization and scale? What are the broader con- straints and possible incentives? TO WHAT USES CAN SOLAR ENERGY BE PUT? The most effective uses for solar energy are in space and water heating and in providing heat for industrial processes. Solar air conditioning is less well developed but could become commercially competitive over the next decade or so. Solar generation of electricity is a far less effective use; although technically feasible, it is expected to be much too costly to compete in the market for some years. Our estimates of the extent to which solar energy could replace non- renewable sources are extrapolations from the CONAES Demand and Conserva- tion Panel's (1976) scenario B (which depends on a quadrupling of energy aA substantial portion of this section is from material developed by W. Harman of Stanford Research Institute as a part of a report to the Energy Research and Development Administration assessing the impacts of the ERDA Solar Energy Program (Stanford Research Institute, 1976). We thank Mr. Harman for his kind permission to use this material. Our solar energy discussion reproduces cost estimates from the Solar Resource Group (1976) of CONAES and from the Demand and Conservation Panel (1976). °The timing of the peak in oil and gas production on a global basis depends on exploration rates, total global resource base, and the rate of growth and demand not only in the United States but throughout the world. Such analyses have been carried out by various groups, of which the early work of Hubbert (1974) and the recent analysis by the Workshop on Alternative Energy Strategies (1976) provides a good overview and anal- ysis of the potential consequences for the world. An energy future for the United States that is constrained to minimum demand growth for energy extends the amount of time available before a transition to renewable energy forms.

39 prices between 1975 and 2010). Tables 5 and 6, later in this section, are summaries of the two differing estimates. Like the CONAES Panel, we assume that the GNP will grow at an average rate of 2 percent per year through 2010. Buildings and industry are the two sectors of the economy in which solar energy can have an impact. Buildings Sector In 1975 there were 70 million residential housing units in the United States; we assume that in 2010 there will be 133 million units, an increase of 63 million. Because of demolition, new housing starts between 1975 and 2010 will in fact be greater than this: 93 million units. Thus 70 percent of the units that will exist in 2010 do not now exist, and only about 57 percent of the units standing in 1975 will exist in 2010. The thermal integrity of the new units is improved over that of existing units. By 2010 new residential units will use only about 75 percent as much heat as present units; commercial, educational, and government buildings will use 40 to 60 percent as much, depending on building type. Starting in 1980, solar systems will begin to be installed, and by 2010 solar systems will be installed in 75 percent of new buildings. After 1980, new buildings will comprise 65 percent of the total build- ing stock. On the basis of linear growth in GNP, we assume that solar systems will be installed in half of this 65 percent, or 32 percent of all buildings. We further assume that solar energy can meet 75 percent of the total demand for space and water heating and air conditioning in these buildings. Scenario B (Demand and Conservation Panel, 1976), on the other hand, assumes that solar space and hot water heat, as well as solar air conditioning, will not come into wide use as quickly. By 2010, 10 percent of new air conditioners, 25 percent of new space heaters, and 50 percent of new water heaters will be solar. Scenario B projects for the buildings sector that total energy use from all sources in these categories will be 25.1 quads by 2010. Only 3 percent of that total, or 0.9 quad, is considered to be the maximum contribution by solar energy. In our view, however, based on the above-stated assumptions, solar energy can meet 75 percent of the total 25.1 quad demand in 32 percent of the new buildings and can thus displace 6.0 quads of con- ventional energy sources. Table 1 summarizes scenario B's analysis of energy uses in the buildings sector.

40 Table 1 Energy use, in quads , in buildings sector, according to Demand and Conservation Panel scenario B Use Year 1975 2010 Space heating 12.6 16.6 Water heating 2.4 4.4 Air conditioning 1.8 4.1 Other (mostly electric) 7.9 13.1 Total 24.7 38.2 Residential 16.5 22.2 Nonresidential 8.2 16.0 Total 24.7 38.2 Primary energy per , unit of housing stock Primary energy per square foot of floor space 234 million Btu 0.3 167 million Btu 0.2 "TJnless otherwise specified Assumes increase from 70.4 million to 133 million units of housing stock. Commercial, educational and government buildings. Assumes increase from 27 billion to 82 billion square feet of floor space. Source: Adapted from Demand and Conservation Panel (1976) Industrial Sector Energy used in industrial processes accounted for about one-fourth of primary energy consumed in the United States in 1975. Experts consider that the industrial sector can use solar energy for most processes that use heat at low temperatures. Three analyses have been made of the extent to which solar energy can replace oil to provide heat at these low tem- peratures.

41 Battelle Columbus Laboratories and InterTechnology Corporation conducted process-heat analyses for the United States (Table 2) similar to that carried out for Canada by Lovina (1976). Table 2 Process energy use in United States industry in 1975 Temperature range Batelle/Columbus Laboratories (Process analysis) quads percent per year InterTechnology Corporation (Statistical analysis) quads percent per year <100°C (212°F): Total 0.17 2 0.26 3 Hot water 0.07 0.12 Direct heat 0.1 0.14 100-177°C: Total 1.34 17 3.25 32 Steam 1.2 2.6 Direct heat 0.14 0.65 >177°C (350°F): Total 6.35 81 6.53 65 Steam 0.5 0.56 Direct heat 5.85 5.97 99b 10.04 Total 7.86 100 aThese figures include about 80% of all process heat; about 20% is in unexamined small industries with unknown temperature distribution. Total does not add due to rounding. Source: Adapted from Lovins (1976) Table 2 shows that in 1975 between 17 and 32 percent of all process- heat energy identified was used at temperatures below 350°F. Thus about 6 quads of energy now used in industry at low temperatures might be pro- duced by using solar technologies. The CONAES Solar Resource Group (1976) carried out an analogous in- vestigation (Table 3) and concluded that 9.5 quads of process heat used in 1975 could be supplied by solar energy. By 2010, they estimated, solar energy could be used for 12 quads of process heat. However, they estimated a high 76 quads for total industrial-sector use.

42 Table 3 Projected U.S. total and process energy use, in quads per year Use Year 1975 1985 1990 2000 2010 Total energy use 75 98 112 146 190 Total industrial use 30 39 45 58 76 Total process heat 19 25 28 36 48 Steam/hot water/hot air for solar-favorable industries 9.5 12.5 Steam/hot water/hot air for solar-favorable industries in "sunshine" areas 4.7 6.2 14 18 24 12 A growth rate in all categories of 2.7 percent per year over 35 years is assumed. Source: Adapted from National Research Council (1979) The CONAES Demand and Conservation Panel (1976) also examined pro- cess heat in the six leading industries (Table 4). Of the total 19 quads used in these industries in 1975, 9.8 quads are for low-temperature steam and hot water, of which 8.1 quads are candidates for replacement by solar energy. In other words, solar energy could have replaced 12.5 percent of U.S. industrial energy usage in 1975. Since the industrial sector is expected to grow more rapidly than other sectors, a corresponding impact of 8 quads in 2010 appears to be a conservative estimate. We conclude that, with the support of a vigorous governmental policy, solar energy could replace 8 quads of oil in producing industrial process heat in 2010. Biomass The conversion of biomass to forms useful for meeting United States energy needs is extremely promising but singularly difficult to evaluate. Among the estimates are these from the CONAES Solar Resource Group: municipal wastes (1.7 quads), agricultural residues (3.5 quads), terrestrial energy

43 Table 4 1975 process heat use among six leading industrial users Type of industry Total use (quads) Percentage in steam/ hot water Energy in steam/hot water (quads) Primary metals 4.2 21 0.9 Chemicals 4.0 60 2.4a Petroleum 1.9 35 0.7 Stone, clay and glass 1.7 8 0.14 Paper and pulp 1.7 85 1.4a Food processing 1.3 ~90 1.2a All others 4.2 ~73 3.1a Total 19.0 9.8 a Source: Data from Dow Chemical (1975); Unger (1975) crops (3.4 quads), and marine energy crops (1.1 quads). Conversion tech- niques include anaerobic digestion, pyrolysis, hydrolysis and fermenta- tion, and direct combustion. Methane, methanol, and other gaseous or liquid fuels can be produced by all these processes except combustion. The combined impact of biomass conversion for 2010, according to the estimates listed, is 9.7 quads. Because of the uncertain state of many of the technologies, we estimate a conservative 3 quads from all biomass sources. Solar Electricity Electricity production using solar technologies is the most uncertain of all. Electricity generated by solar thermal plants is currently esti- mated by some to cost 30 to 50 times as much as coal- or nuclear-generated power. The best estimates developed by the Solar Resource Group (1976) for the cost of solar electricity generation by 2010 are $0.042/kWh, whereas conventionally generated power costs only $0.03/kWh. However,

44 the figure of $0.042/kWh is quite inappropriate as a cost estimate for the extremely complex and cumbersome approaches to solar electricity production using solar thermal techniques now being developed by the U.S. Department of Energy. At best it is an optimistic figure, even for simpler solar electric technologies. On the other hand, one must be careful not to prejudge the oppor- tunities for innovation. A solar thermal design by Otto Smith of the University of California at Berkeley differs in fundamental ways from those concepts being pursued by the Department of Energy. Professor Smith believes that busbar electricity costs as low as about $0.036/kWh are feasible, even including enough storage to permit day/night load leveling. Further analysis and testing are clearly in order. But even to proponents of the wider use of solar energy, solar elec- tricity does not appear to be a feasible competitor to other sources of electricity generation. The only way in which solar electricity could play a significant role by 2010 is through vigorous and massive federal subsidy or through a policy that would not permit development of other generation sources, as in the event of a nuclear moratorium. Neverthe- less, widespread interest in solar electric systems will probably encour- age the government to build a number of solar thermal plants, virtually regardless of cost. If photovoltaic devices were to be developed vigorously, they could become important competitors to conventional fuels. Many of the mate- rials, such as silicon, that are sources for photovoltaic devices are intrinsically cheap. The difficulty lies with the processing. There is every possibility that, once major advances occur in solid-state physics, application of those sophisticated techniques will lead to mass produc- tion of photovoltaic devices at prices not significantly above the costs of window glass. Indeed, this has reportedly already occurred for the low-efficiency material cadmium sulfide. Thus, innovative research seems required before costs can be lowered. Under these circumstances, the potential impact of photovoltaics seems impossible to assess and certainly defies meaningful economic analysis. In brief, major allowances for solar electric systems do not now appear appropriate, although because solar thermal electric plants will probably be built, we allow for solar electric systems of all forms to replace 2 quads of primary fossil or nuclear fuel. This would be the equivalent of about 36 gigawatts of conventional generating capacity op- erating 60 percent of the time (that is, with a 60-percent load factor) and would amount to about 6 percent of installed electricity production In 2010. In summary, solar energy could produce 6 quads of the energy esti- mated in the CONAES Demand and Conservation Panel's (1976) scenario B to be required by the buildings sector in 2010. It could replace 8 quads of the energy estimated to be required by the industrial sector and 2 quads as solar electricity. Biomass could replace another 4 quads of energy from nonrenewable sources. These 20 quads of energy from renewable sources compare favorably with the 1.9 quads from renewable sources estimated by the Demand and Conservation Panel (summarized in Table 5), which is the base case. Table 6 shows the change in the base case with the energy input in the Solar Modification Scenario.

45 Table 5 Projected 2010 primary energy input in quads, from Demand and Conservation Panel scenario Ba Energy source Use Total Buildings sector Industrial sector" Transpor- tation sector Coal 0.07 26.0 0.2 26.3 Oil 4.8 23.0 25.0 52.8 Gas 5.1 5.7 0.2 11.0 Electricity 26.8 8.9 0.2 35.9 Solar (0.9)C (1.0) - (1.9) Totald 36.8 63.6 25.6 126.0 JPrimary demand is calculated from end-use demand by converting oil at 1.26 Btu of input per Btu of output energy, gas at 1.16 Btu of input per Btu output, electricity at 3.1 Btu input per Btu output, in accord- ance with Bureau of Mines convention. 8.82 quads of agricultural and construction use transferred from buildings sector to industrial (Clark Bullard, U.S. Department of Energy, personal communication). £ Solar use is actually greater than figures shown due to use of passive solar design. However, it is impossible to separate passive solar from overall energy-conserving design of building. d Excluding solar Source: Adapted from Demand and Conservation Panel (1976) HOW SOON CAN SOLAR TECHNOLOGY BE AVAILABLE? Plans for introducing any new technology are usually considered in two phases. Start-up time is the year in which commercial prototype units are first installed. The systems constructed may not themselves be economical, but there are sound engineering data to show that, under conditions of mass production, the systems will become economical. Take-off time is the year in which full-scale commercial production begins. By the time production begins, system reliability has been assured through field testing in the interval since start-up time and

46 Table 6 Projected 2010 primary energy input in quads, with solar modification3 Energy source Use Total Buildings sector Industrial sector Transportation sector Coal 0.07 26.0 0.2 26.3 Oil 4.0 16.6 25.0 45.6 Gas 4.3 4.1 0.2 8.6 Electricity 22.4 8.9 0.2 31.5 Solarb (6.0) (8.0) - (14.0) Total (nonsolar) " Q d 30.8 55.6 25.6 112.0 Demand and Conservation scenario B*. Primary demand is calculated from end-use demand by converting oil at 1.26 Btu/Btu, gas at 1.16 Btu/ Btu, and electricity at 3.1 Btu/Btu, in accordance with Bureau of Mines convention. Excludes solar electric and biomass. Solar is assumed to displace oil and gas only for process heat. f> Solar heating and cooling substitutes virtually interchangeably for all energy forms, according to the Demand and Conservation Panel analy- sis. In scenario A essentially all of the energy requirements for heating, cooling, and hot water could be replaced by solar (14.6 quads). Solar electricity (2 quads) and biomass (4 quads) may be allocated arbitrarily among the demand sectors. Biomass replaces gas or oil. Source: Adapted from Demand and Conservation Panel (1976) system costs are convincingly known to manufacturers and are competitive with conventional energy forms either because of intrinsically low costs or because of price guarantees, subsidies, or other support. Table 7 presents estimates of start-up and take-off dates for each major solar technology. These start-up times are believed to be achiev- able through a vigorous research, development, and demonstration program. The take-off times are the times for critical decision making, and national action may be required to assure that the deadlines are met. Installed capacity at start-up time is at least 0.03 quad/year, delivered, for solar systems that produce electricity, and 0.1 quad/year

47 Table 7 Start-up and take-off dates for major solar technologies Technology Start-up date Take-off date Heating and cooling Active 1975 1980 Passive 1960 1975 Biomass 1985 1995 Solar electricity 1985 1995 for other systems. One billion square feet of solar collector corresponds to 0.1 quad capacity per year. Installed capacity at the time the take-off point is reached is assumed to be 0.5 quad for heating and cooling systems and 0.2 quad to displace primary fuels for central-station electric plants. If the electricity-generating systems were decentralized and used photovoltaic devices, the capacity required would be substantially less. After take-off time, the various solar technologies are assumed to grow at the rates shown in Table 8. These growth rates will moderate as the market becomes saturated. However, except possibly for domestic and commercial water heating, market saturation is not likely to occur before 2010. Table 8 Projected growth for solar technologies Application Energy produced in quads per year Growth per year (percent) 1980 1985 1995 2010 Industry 0.5 8 12 Heating/cooling 0.5 6 10 Biomass 0.5 4 9 Electricity 0.25 2 15

48 20 15 M Q < 10 o Electricity Biomass Heating/ Cooling Industrial Process Heat 1980 1990 2000 YEAR 2010 Figure 3 Estimated growth of solar technologies, 1975-2010 The CONAES Modeling Resource Group (National Research Council, 1978) estimated the limit on growth rates for new technologies at 20 percent per year. Our projected growth rates of 10 to 15 percent per year are well below that estimate and are also modest compared with most projected growth rates for nuclear technologies. On the other hand, it is common to delay the introduction of a technology. There can be no question that these growth rates are plausible only if vigorous governmental policy is assumed. Costs for solar electricity are expected to remain prohibitive at least through 2000, barring a major technological advance. Some solar electric capacity will, however, surely be installed if we are vigorous about developing solar energy. Nevertheless, in contrast with solar heating and cooling and solar process heat, solar electric generation cannot now be considered economically feasible. Subsidies will be needed and are assumed to operate. It must be recognized that solar energy installations come in many sizes and shapes and have extremely diverse potential. Some of the technologies will surely fail, and others will take forms quite different from those now envisaged. For long-range planning purposes, such shifts have little importance. It is the total impact that matters. Figure 3 illustrates the combined impact of the several solar technologies.

49 HOW MUCH WOULD SOLAR SYSTEMS COST? No cost numbers for solar energy systems are credible—at least not within a range that could make decisions on solar and nuclear systems clear and unambiguous. The decisions facing our society as we move from the fossil era to the era of renewable resources transcend cost considerations. Far more important are the implications of the energy systems chosen for availability of energy, social structure, individual freedom, and relations of the United States to the developing nations. It is to these issues rather than to cost that we urge the reader to address her attention as she reads the remainder of this section. The Solar Resource Group (1976) of CONAES, under the assumption that major research and development programs are successful, estimated that by 2010 costs for delivered solar heat and delivered solar air conditioning (Table 9) will be comparable to or below conventional fuel Table 9 Delivered solar energy costs in the year 2010 Delivered energy Cost in dollars per million Btu Domestic water heat 4.40 Passive space heat 4.60 Active space heat 7.30 Residential air conditioning 19.60 (6.1) Nonresidential air conditioning 5.00 Process heat 3.76 .oob Bioconversion 0.40-9 Electricity 12.40 (A. 2) a Cents per kilowatt-hour in parentheses Depending on source Source: Adapted from Solar Resource Group (1976)

50 costs alone for scenario B of the CONAES Demand and Conservation Panel (1976), shown in Table 10.a The argument on behalf of solar energy is even more favorable when equipment costs are taken into account. Thus, from this point of view, a discussion of solar energy's role depends only weakly on economic factors; it depends mainly on market penetration. (The solar elec- tricity situation is a bit different. There are great questions about costs, reliability, and energy-storage technology.) The point is that, even for sources of supply whose technologies are more or less at hand, the costs of entire systems have not yet been calculated; nor is it usual in energy analyses for such calculations to be made. Government planning documents, such as ERDA-48 (U.S. Energy Research and Development, 1975) and ERDA-76-1 (U.S. Energy Research and Development, 1976), include only costs related to supplying the energy, Table 10 Energy cost assumptions under Demand and Conservation Panel scenario B, in dollars per million Btua Energy source Year 2010 1975 Distillate 6.74 2. 81 Utility residual 4.85 2. 02 Natural gas (total) 7.74 1. 29 Electricity Commercial 19.04 (6.50) 9.52 (3.25) Industrial 10.72 (3.66) 5.36 (1.83) Residential 18.22 (6.22) 9.11 (3.11) Demand-weighted average 15.82 (5.40) 7.91 (2.70) Cents per kilowatt-hour in parentheses Source: Adapted from Demand and Conservation Panel (1976) aThe dramatic oil-price increases announced by OPEC in 1979 make near- term solar economics look far better than they did only two years ago, when this report was first drafted.

51 not those associated with its use. Thus there are estimates of the costs of generation and transmission systems for electricity but not for the costs of heat pumps when electricity is to be used for heating. Such partial analyses of costs, which stop far short of comprehen- sive systems analyses for these familiar energy sources, would require expansion to total-cost analyses when systems using solar energy, which combine investment in conservation with investment in supply and less- tangible costs for health, aesthetics, and so forth, are being considered. Indeed, a nationwide analysis of all aspects of the entire energy system, from source to use, including insulation, lighting levels, heating, ven- tilating, and air conditioning systems, and others, would necessarily emphasize energy conservation by focusing attention on the efficiency of the system as a whole. Once a comprehensive systems analysis is undertaken, however, the extent to which solar heating and cooling can cut our use of nonrenew- able sources will be apparent. Only then, perhaps, will government plan- ners recognize the contribution possible from solar energy. In brief, a systems analysis would enable us to see that the effi- ciency of the system requires us to conserve nonrenewable sources of energy and thus to turn to renewable sources wherever possible. An econ- omy based on the fullest use of renewable energy would therefore lead us toward a future in which less energy will be used. A comprehensive systems analysis will also be favorable to solar energy from the standpoint of forever reducing our dependence on foreign oil and foreign fissionable fuel for generating electricity. One comprehensive set of supply costs has been developed (Lovins, 1976). For example, Lovins finds the total cost of nuclear systems, taking into account fuel cycle costs, transmission and distribution costs, and a number of other factors, to be close to $5000 per kilo- watt of delivered power. This is so much larger than the cost of a plant alone that further examination is clearly in order. Lovins1 sum- mary of the costs of delivered energy using a variety of forms appears in Table 11. Investments in energy conservation are frequently far more cost effective than expenditures on supply expansion. Also, many services are performed better when energy use is decreased (for example, auto- mobile air pollution is reduced when cars use less gasoline per mile). Thus it is proper and desirable to associate a renewable-fuel-based economy with a low-energy future. There are many reasons why it might be advantageous to use solar energy even if costs were higher than for conventional systems. Con- sider electricity. An optimistic busbar cost estimate for solar elec- tricity, including required storage systems, is about $0.045/kWh. This is about $0.15/kWh greater than current busbar cost estimates for coal- or nuclear-produced electricity. (Consumer costs for electricity in some parts of the United States today are running much higher than this—as much as $0.07/kWh.)a The additional annual cost to produce 1.9 trillion a These numbers are typical for 1977, Coal-fired plants scheduled to begin operation in 1985 may produce electricity for as much as $0.08/kWh or more.

Table 11 Capital cost of energy supply and conservation technologies Technology Capital cost (thousands of dollars per barrel of oil equivalent per day) Traditional direct fuel technologies, 1950-1970 2-3 Imported oil or domestic coal, 1970s 2-3 Frontier oil and gas, 1980s 10-25 Coal synthetics and exotic hydrocarbons, 1980s 20-40 Central coal-electric with scrubbers, 1980s 170 Light water reactor, mid-1980s 200-300 Fluidized-bed gas turbine/district heating/heat pumps, early 1980s 30° Wind-electric 200 Retrofitted 100% solar space heat, mid 1980s 50-70° Bioconversion of agricultural/forestry residues, 1980s 13-20 Pyrolysis of municipal wastes, late 1970s 30 Improved end-use efficiency Q New commercial buildings 0-3 Common industrial/architectural leak-plugging 0-5 /•» Most industrial/architectural heat-recovery systems 5-15 C- Difficult, extremely thorough building retrofits 25 a!976 dollars Delivered in the form of electricity f- Including cost of end-use devices to deliver desired function Source: Lovins (1976)

53 kilowatt-hours of electricity, the amount used in the United States in 1974, is then about $30 billion. This is about half the nation's 1979 total oil import bill. A national decision to produce 10 percent of this electricity using solar power—which would mean an installed solar capacity equivalent to 38 conventional plants operating at a 60-percent load factor, would carry with it an additional cost of just $16 per person per year. An even more interesting way to look at an extra cost of solar energy of $0.015/kWh is as an offset to oil imports. The oil-fuel cost corresponding to $0.015/kWh is just $8.80 per barrel (when generation loss is included), which is much less than imported OPEC oil now costs. Thus a premium paid for solar energy can be considered an investment made within the United States to cut back—forever—on oil imports for electricity generation. The investment has a favorable benefit/cost ratio. In short, although we know less about the costs of entire systems of solar energy than about those of systems using conventional fuels, a comprehensive analysis will prove favorable to solar energy on grounds of conservation, of reducing our dependence on foreign sources of oil and nuclear power, and on the acceptability of the higher cost of a por- tion of our electricity's being generated by solar energy if nuclear energy and coal are our only alternatives. ARE THERE NOT GEOGRAPHIC VARIATIONS IN SOLAR ENERGY? Solar energy varies geographically as well as seasonably, and of course storage must be taken into account when any solar use is being studied. However, solar buildings have been constructed to take maximum advantage of solar energy in Scandinavia, England, and Canada, whose climates are in places far less attractive than any in the United States, and they are proving cost effective. In most parts of the United States, solar hot-water heating and elements of passive design are already cost effec- tive, and this situation will improve as costs of conventional fuels in- crease and as more experience is gained with solar energy. CAN SOLAR-ENERGY SYSTEMS AND INSTITUTIONS BE DECENTRALIZED? Technically there are no compelling reasons to make decentralization of solar energy uses an important issue. Economies of scale in solar tech- nology go both ways. Both the centralized solar electric plant and decentralized heating and cooling have significant roles, and neither excludes the other. Decentralized solar electric power, particularly with advances in storage and photovoltaic techniques, has some advantage from the standpoint of system resiliency. Further research is required. From a technical standpoint, considering factors of economics, system performance, aesthetics, public safety, wartime vulnerability, and mul- tiple use of sunlit areas (as with collectors on rooftops and over high- ways) , applications will arise in which on-site solar technology fits and

54 others in which centralized solar electric technology is better. However, these criteria alone are not likely to decide the outcome. Decisions will be influenced also by the issue of decentralized control. The symbolic power of the decentralization issue should not be underestimated. Solar energy is democratic; it falls on the rich and the poor, the weak and the powerful. To decentralize solar energy use is to enable individual homes or communities to gather it. The issue of decentralized solar power is symbolic of a greater issue: the pre- servation of liberty and equity by maintaining some independence from the "big system." As the theorists of free-enterprise democracy con- sidered the principle of control and ownership of property essential to liberty, so the principle of control over indispensable energy supply is now being put forth as a precondition of liberty. The Symposium on Future Strategies for Energy Development held at Oak Ridge, Tennessee, October 20-21, 1976, brought into sharp focus many of the issues involved in contrasting dispersed technologies with con- centrated ones. The arguments for "soft" technologies were presented by Lovins (1976) and those for the capabilities of concentrated tech- nologies by Haefele and Sassin (1976). Lovins described the philosophy of small-scale systems: Some diseconomies of large scale are starting to be widely appreciated. For example, it is now well known that large electrical components, notably turbogenerators, often lose in reliability—hence in contributions to grid operating costs, standby capacity costs, grid instability, and lost revenues—what their size gains in unit capital cost. These effects are so common that there is mounting evidence that most types of power stations can have lower busbar costs in sizes of the order of hundreds rather than of thousands of megawatts. Dispersed generation near load centers is well known to improve system integration and stability. According to one recent study ... it was found that one kilowatt of dis- persed generation was equivalent from the standpoint of reserve requirements to 2.5 kilowatts of central generation. The reliability of supply within the network was determined by means of an index related to the LOLP (loss-of-load prob- ability) . The reason why the dispersed device can be so effective is that it protects the load in its vicinity against generation as well as transmission and distribution outages. Thus the "dispersion credit" traditionally as- signed to local supply—e.g., from battery banks or fuel cells associated with distribution facilities—may be far too low, since it reflects only the cost, on the order of $100/kilowatt, of saved transmission facilities. Unfor- tunately, publication of a study by the Electric Power Re- search Institute with this conclusion has been suspended, so few recent data are available. The extreme capital in- tensity of all electrical facilities offers ample reason to

55 explore carefully the reliability implications of more cen- tralization. The classical literature of this subject seems to me sketchy and unpersuasive and its quantification prim- itive. A further diseconomy of large scale is so obvious that it is often forgotten. The past few decades' military experi- ence in Europe and Indochina has taught us that central energy systems reliant on a few large facilities are far more vulnerable, and harder to restore when damaged, than dispersed systems. . . . Small energy systems suited to particular niches can mimic the strategy of ecosystem development, adapting and hybridizing in constant coevolution with a broad front of technical and social change. Large systems tend to evolve more linearly like single specialized species (dinosaurs?) with less genotypic diversity and greater phenotypic fra- gility. Large systems also accrete costly, specialized infrastructure that strongly influences future lines of development. Thus unamortized natural-gas pipelines, the third largest U.S. industry, provide a strong incentive to make synthetic pipeline-quality gas even at a price an order of magnitude higher; building a grid dependent on 1,000-MW blocks of electricity discourages a future shift to smaller-scale or reduced electrification. Small sys- tems, in contrast, tend to depend more on infrastructure installed at the point of end use, thus increasing the user's ability to adapt. For example, resistive heaters and electric heat pumps are not very adaptable, so an all- electric house is hard to heat except with electricity from some source. A domestic or district heating system based on circulating hot water, however, can use virtually any heat source at any scale without significant change to the domestic plumbing, adapting to as wide a range as solar collectors, solar/heat-pump hybrids, and combined-heat-and- power district stations. Thus if a transitional technology such as coal-fired fluidized-bed gas turbines with district heating is deployed first in those urban areas where solar backfits will be slowest and least convenient, with dis- trict heating clustered in holding tanks of neighborhood scale, then that interim heat distribution system (coupled to existing domestic plumbing) can be adapted later to whatever soft source of heat becomes available in each neighborhood. Infrastructure at or near the point of end use can be designed for this sort of piggybacking, whereas large-scale distribution infrastructure enables one only to choose one or another kind of enormous central power station, gas plant, etc. Haefele and Sassin note that laws of scale tend to drive technol- ogies toward increasingly large units but that on the other hand the

56 need for redundancy drives in the other direction. For solar systems a major problem is the need for backup systems and for storage, both of which are expensive. Day-to-day energy-storage problems are probably surmountable, but storage to keep pace with seasonal and annual vari- ations is far more difficult to achieve (or requires biomass as the medium of storage). One approach is to disperse the system broadly but to give it the capability to move extremely large amounts of energy from one region to another. A study by R. Partle for the International Institute for Applied Systems Analysis (IIASA), reported by Haefele and Sassin (1976) analyzed movement of 10 gigawatts (electric) over dis- tances of 4000 kilometers, estimating a unit' cost comparable with that of moving coal by train. Marchetti (1975) at IIASA has explored the concept of the energy island, which might use breeder reactors at a level of about one tera- watt (1000 gigawatts) of thermal energy. This amounts to about 14 per- cent of the energy used by the. entire world today. The concept is intriguing, for it represents an extreme case of decoupling the tech- nology of energy supply from the users of energy. Whether such decou- pling could decrease the anxiety associated with large-scale nuclear systems remains uncertain. What is clear, however, is that there are extraordinary choices to be made in terms of energy-system scale and decoupling. In the final analysis, the issue of large versus small is the issue of coupling people to their energy-supply systems, not an issue of solar power contrasted with nuclear power. Nuclear systems could be made small and used for both district heating and electricity production. Solar systems can be made vast, sited in remote areas with high inso- lation, and coupled to energy users through massive electrical networks or through other transmission devices, such as liquid nitrogen gas, methanol, or liquid hydrogen tankers. THE ENERGY POTENTIAL OF SOLAR: OTHER VIEWS Any attempt to analyze the energy potential of a new technology is fraught with uncertainty and subject to challenge from many fronts. The problem is especially severe in solar energy, for experience to date is so limited and the institutional barriers to success are so great (Schoen, Hirschberg, Weingart, and Stein, 1975). The variations possible are ex- hibited in the broad range of estimates of solar impacts compiled by the CONAES Solar Resource Group (National Research Council, 1979) and sum- marized in Table 12. Clearly, there exists no consensus on how to estimate the potential impact of the solar technologies, even on the assumption of highly com- petitive prices, when the range of impacts by the year 2000 varies from as low as 1 quad to as high as 94 quads. Thus the Solar Resource Group estimated the potential for solar heat and electricity. Figure 4 expresses solar-energy contributions as a function of energy cost in 2000. Figure 5 expresses the impact of solar technologies over the period 1975 to 2010 for specified costs of $5/million Btu for heat energy and of 60 mills/kWh for electricity.

57 10 "~ Solar Energy Supplied as 10 Solar Energy Supplied as Heat, Year 2000 Q Electricity, Year 2000 LU Photovoltaic — »> o O LU 3 cc UU — LU a. Ocean Thermal < — _J OE Energy Conversion LU Q — LU D - < - — -"5 I Wind LL. 0 O Active Space u_ — IO Heating ^^^ "- Q ^ ~~ ^t^^. m O Process Heat <£ Q Systems So|ar Therma| Electric Non-Res. Air-Cond. ^T ^~ 1 ^ Passive Space Heat 1 ^ — Domestic Water Heat Wind for Interruptible Loads O 1 1 1 1 1 1 1 1 n I 1 1 1 1 1 1 1 02468 COST OF ENERGY, $/million Btu (1975 $) Figure 4 0 20 40 60 80 COST OF ELECTRICITY, mills/kWh (Busbar, 1975 $) Estimated energy contributions of solar energy technologies in the year 2000 as a function of costs of competing fuels (National Research Council, 1979) The CONAES Demand and Conservation Panel (1976), working in con- junction with the Solar Resource Group, approached the problem by as- suming that solar systems will penetrate the market gradually once they become competitive. The procedure uses a functional form for the com- petition, thereby assuring that the new technology occupies its place gradually. This approach leads to a very low impact in 2010—just 1.9 quads in scenario B. However, by 2010 solar technologies will begin to have a fairly rapid impact. Figure 6 portrays the impact of one kind of solar technology: residential space heating. Another approach to estimating the impact of solar energy is sug- gested by Lovins (1976), who analyzes energy use in terms of the required thermodynamic quality of the energy. His approximate analysis for Canada is shown in Fig. 7. The figure compares 1973 end uses with 2025 end uses

58 rH ItH rH • O (0 vo rO ^3- rH r>. u 4J ^3 1 00 ^D ^J^ i <T> CM O m • i i CO • I oi c H CM • sr vo st O 00 a o rH m CM co 1-1 •* co U-l vO rH rH m iH CO .u CU CO CO g ^ ^^ o cd i4H rH 0 rH 0 0 < H co • vO rH M 1 r- • CO CU m m m CM o\ CM O ,£3 oo I 1 r~ -H £J vO CO <f rH rH CM 00 in m vo vo 0 0) CU -^ • H 3 00 •H rH er cfl 00 ss d CU rl 0 CU O •H O CU O EC "H 0 0 -H 4J * oi co l-» 0 • (d o •o n CM •* O £5 o 0 rH CU oo cy\ o CO cfl > e c m rH o\ • i t CM T3 C O CM O rH rH O l-H rH O O rH \O r— 1 -• M 0 CU U 'O cO CO •H .fi c > IS *J cfl >> m CU •> -x3 CU rH A co oi 4J rl -H O O MH 0 •H •fH rH CO f-< cfl 4-1 rH • -H CO S rH C 0 < M D. a \O o CU > oo r^ 1 • OT •H J2 0 rH CO rH CM CO rH o\ rV U 4_J 4_1 1 1 • • t CO 0 CO o m r- o IH vo O rH CO \O •S • O •H >, CU H cfl CX rl U rH rl rl P. o -a CU 3 cfl" w c5 60 CO 18 PS iS rl CU m - & 0 CO -* i-- m m rs M CU CM r^ CO fo <f P^ ^f r — r^ in CU M CU 4J CU •H CU CU •^r s CO 4J O cx > > M 4J > rl | 0 00 rl rH O 0J CU O O cfl CO O CO rH Q O M £; 9B X rH Z 3 cd co T3 0 O O 14H O •H O IS 4-1 co 1 J4 cfl u a rl £2 CT» CO oai oo CU ca m cfl rl >% VH 0 c •> vo u cd u -HW0 o B • CO rH CU CO > •H O >» -H •H M CO CM rH 00 £3 rH -H 00 4J s CM 0 Cd C W CO 4J rl CO .H 3 CO CU rl -X3 PM bo bo. w 4J^3 U 0O*T3c3iJ CU W cd O 0 W co § M PM . CU v-x CM u oi rH g -H C C rH 0COJ3 rHcdOrH0 -d 00 CTi CO cfl W CU CU cfl.<; T3 •H4Jfii cfl-H rl -sT ->T rH > 0 CU U >^yrlWO|3 MB O<! COPQ CCUTJ CUTS CU 1 1 CU rH <3 < T3 rH Q Q 0 CU D5 PS cfl O rl -H rH Cfl iH 0 CO rH H 30 rl X><! C cfl > O rH 1-1 0 O •< O H O O CU Q O CU H CO Q 55 C/3 g U (xi PL, H U W U cd co co

59 §1 0 rl U M 4-l • CO >><-v >, 00<f rH • • g M 4J cfl CO cd 0 • rl CU 4-1 CU -H cu » co CU CO .0 CU . en o r-- cd co 0 O 0 CO 0 .0 CU 4J 3 u rH \O o (U W CX O CU -H O pe< CO 4-1 CO O CM - -H 4-1 4J £ cu •H CO PS S 3 IH CO >. IH ca O H co M 0 rH PH CU CU J2 1 0 3 H rt CO CU 00 IH •a co o •* cu cd M 4-J CU •H 00 CO IH «4H c 0 CU r- S 0 SH •• u rt rl 4-1 CU CU CO O M •H H 1 ex o a £>, IH g CU 0 pH* 0 O 00 Z O ^ cd ,0 O • O 0 CU U4 iH m a 0 IH ^ rH 4J •• j_j fa ^«^ U u B ^ M W tfl CO CO • 00 4J r^ CX 00 • CO ^ "H rH CU > < cd o M >s> <~> iTj l-i O IH CO 0 ca 9 rH O O CU ^ rl O co m JH O rH CU * CU cfl 0- cfl en -H QIB . PH ex cu CO OO H -3- 4-l 4H CU 0 t3 0 H •H rJ 0 > „ cn •H 14-I S-i « CO CU a 0 PQ CM Ln 0 CU rl 0 CU .0 U cd O iH 1 O cfl Q M •• • U 4J en O CU •H cu •H iH cfl W rH cfl • -• ^i •H O CO IH O 14H hJ 4J cd rH PH T3 rH CJ 0 B 4J CU 0 0 CO 0 14H u cd 0 . 1 ^^ /• cu • 0 o • 0 0 CU rH O O co co O ^JT] •H fa 0 0 rH cfl co o •rl 0 W -H -H CO •H rl pM CU CO O cd 4-1 O 4_] 4-1 4-1 CU O 60 CU « o PH -H 0 .0 rH •> CX 0 rj IIH cfl tfl PH cn 0 • CO 4J -H 4-1 o o §0 g W r.^. * DOS •H CO 0 • . 0 > ^ tfl •H |H 0 rj CO /-N •H rH >-, 0 4J £ O CU M •H M 00 rl 4J CO O 4J CO 0 rH >» rH O 00 O 0 cfl CJ . M CU IH CU rl O cfl CU •H 00 0 cfl •O 1 4-i CX MH Vi lH IH cu 3 00 ex cn cu cx Z co 4-1 (3 O O CU CM CU (X 0 CU CU IH 60 >l O 4J 0 cu 0 rl CU cfl -H CJ -H > m •H <J IH 0 CU PH O 00 -H 3 -H 3 0 W 0 Z »0 rl •H f-~ O U 4J J^ M UH fa i — 1 rH 0 CJ CO >^ CU rl rH O rH rH 4-1 4-1 fa (U UH rH PQ -H 4-1 cu >s cfl 00 S cu I en cfl to rl TH 0 0 O >^ cd f. >-. . 00 rl <3 Q W •H 0 cfl B cu >^ py 00 PQ CU CO 4-1 IH cfl M JZ CU 1 fa 0 rl iH rH E: 6 00 00 M O S 3 m cu 0 CU ^J ^^ 000 ^4 rH 0 CU 0 rl O 4-1 r- 0 r^ • cd U co j3 en cj IH cu CO -H 0 •• CU O 4J 0 cr. cd ON 4J 4-1 o cu •H CU O cu 0 0 4-1 W • •O MH CU rH rH M 73 CJ 0 rl IH CU cu - > H O 0 CO 4-1 CJ O 0 0 •H JH H X >. 0 •H -H CO CO CU H IH CO a CO *H cfl > cu 4-J 00 O M 4-1 IH - cd ex O *& • - . cu IH H rH rl cd CO PH cd X CU rH CX IH 0 i — * 0 Pi CU CO 4H CU cd 0 cu • rH ' Z 0 13 cfl CU -H o o rH CU o co T3 0 O 0 CO O 4J •H •> 0 0 PH" r0 •H •H 0 >•> u 0-0 w • CO ^ 0 M CU M vH H 4-1 -• 4J 0 4-1 3 0 0 co CD CU CU 00 rj CO • cfl -H CO 4-1 O O 00 0 0 0 B IS • CU rl U 4J * IH 4-l 4J -H CU 3 •H -H 4J 4-1 0 H CO o O O >4-l fa 0 CO Z •H .-T W rl CU CU O tr^ CO Q CO 0 •H CO to I-i CU • CO r~ 01 4J -' — ) E~i -H •H •H -H 1-3 rH N B 4J re to cu cj •H •• B O 4J 0 « 0 -4-1 CO •H IH CO PH •• CU 4-1 • CO rH O CU cfl ^ iH O •• -s o •H CU O rH O 'M W • SC 4J Q o CO CJ PH CO ^3 j1^ B P ^ . CJ cfl m s*l 4J H efl •H rH o 0 M <i 00 m •H 0 o\ CO > B " cd • • 0 O M O <! a\ r~- cu U M 1^ •H • t- 0 0 >, CO . -H O ^i 0 4-18^ CTs ^ IH ON c . « vr O o 43 4-1 • 0 0 <r 4J LH 4-1 -H rH -H CU rH rH •H cn 73 r^ u 4J •H t3 o r^ co jj 0 r0 0 CO <J 4J B CO S r^ rH CTv ,O 00 CO rH 0 -H o\ Z 0 CU CU CO CU M Q cd B cj 73 CTi CU < rH -H rH 3 0 § CO 4-l 4J rH cfl Z S 60 K • 0 IH IH 0 •H CJ 0 rH en •H r-5 -- cfl rH &s ex o u •H IH (~! 4-J O1* • CU 73 •• PH o O rl ^ CU CU J3 U CU 00 CO co CJ B 0 • • <! co • H •H PH 0 4-1 T3 0 0 0-0 co * cd ^>, rH • •H 3 0 cfl cu • CU 0 CU rH 0 CU 3 CU CM IS co cfl Q H 0 O > 0 > 0 > P 0 •H <! CO H 0 ex id IH > m 4J •H cfl CU O CU O -H S 1 U CO M (X •H O\ •> 0 « ^ 4-l » .. O Q -H P -H 4J 4-1 rH oo co 4-100 g CU C r^ cfl 13 >. -H 4-1 4J CO * m 0 Cfl r- T3 O CO rH O m B o Vi M rH 00 IH T3 CO T3 cfl IH l-3 CM CU 0 U 0 Vi 4J •H r.^ 0 4-l • O 4-1 CU IH CU 0 rl 0 H 4-1 O B o rl td PH • 4-1 0 • CO rH /-A O 00 <r fa CO -H SI CO 4-l CO tfl 4-1 CO CO -H • CM CX -H O 4-1 cu pi co rl 0 •H 14H CU CO >> IH to . at cn •H •H a\ cu 0 00 W ,0 0 J3 0 0 T3 rH tfl CO 'J 4-l 0 cu W to rH ^ r0 rH ,0 •H 0 IH O O O O -H 73 0 cu Z cu •H rH O. . m 0 CO 4-1 S -H O rl B rl B S 0 CO I PC; 4J 3 CU ex •> 4-1 0 \D W cfl 73 rl * 4-i cfl CU CO CU TJ cd oi 3 0 ps1 CU 4-1 CO O 1 S • M-J < ex m CU O CU Q < M Q •• rH cfl cfl PS O rH •H PH 0 0 0 en r^. co CO CO - o cd 0 fa rH •^ 4J H O 0 ^, CTi CU CU T3 • CU T3 4J • o CU cfl 0 o cd 14H •> CO S ( •H 4J 60 rH O PS1 0 x-s PH 0 0 "-J 0 4J " 0 M M 0 O 0 CO M s— " rH 0 4J O VH • •H cd oo cd cu CM •H CU 41 -H o co o •H CO O rH cfl CU cu >, o 4-l < | fa 01 .H CO ex o 0 00 00 4J 00 4-1 CX PQ - O 73 cd CO 4J IH M O c r^j 0 0 W CJ rl 0 < M 0 O m 4-1 rH rH 3 « 3 CO 00 0 •H 3 CU CU CU Q CU CU rH « oo CU 03 CO J± 0 to 0 CJ 4J O 4H 0 PH rH 0 « rH 0 B PH 0 B co co a\ CU -H 0 0 33 4J O cd CJ rH fa cfl a rl CO w ex w W CX > rH rH > 14H •• O 0 M CO U H M CO rl cu o o .^ o co cu •H 00 cu •H 00 CO • 0 PH O • JT] T3 O CU IH rH -H • rH • rH Q 73 CO 4J 0 u 4J -H CU CO O 73 H CX en M •H 13 CO rH 4J en co 0 en co 0 rl CO -H rl CO CO CO • B rl M IH - 73 O rH CU rH CU -H • > o • > T3 cfl cfl rH M 3 Z CU CU p co en IS O P c fa O fa 0 H M C3 CU •H ^ CU 0 CJ CU cu ex O 3 Q erf O Q CTv ' T3 CJ 0) cfl i 4H PH 00 CO , a cj •H Q 4J •r-> Q Cfl , 24 >H pi cn M

60 10 r~ Solar Energy Supplied as Heat at Five Dollars per Million Btu 10 Q_ UJ CC — Process Nonresidential Air-Co nditioning — Passive —Domestic Water 2010 M O Solar Energy Supplied as Electricity at 60 Mills per Kilowatt-Hour 1975 1985 1995 YEAR Ocean Thermal Interruptible Wind 2010 Figure 5 Estimated increases in energy contributions of solar energy technologies used to supply heat at less than five dollars per million Btu or electricity at less than 60 mills per kilowatt-hour in the year 2000 (Solar Resource Group, 1976) when energy is supplied by a "super-technical fix" and, alternatively, by a "soft" source. The "soft" energy mix uses solar energy for virtu- ally all low-temperature heat and would thus provide about one-third of the total energy input to the Canadian system. There is no allowance for solar electricity. Finally, the United States government's approach has been to em- phasize solar electricity rather than other forms of solar energy. The President's budget to Congress for fiscal 1977 included $102 million for solar electric research out of a total request of $160 million for solar energy. Congress acted by appropriating $174.9 million for solar elec- tric energy and $290.4 million for all solar research and development. The Department of Energy continues to emphasize solar electricity, though less than in years past. Most estimates of the costs of generating elec- tricity by solar energy are considerably higher than the costs associated with conventional coal-fired or nuclear-fired generating systems. Until recently, solar energy and renewable technologies have not received serious government attention. In part as a result of citizen action, this situation has changed strikingly in recent years, as is evident in the budget for solar research and development. The federal budget has grown exponentially for several years, with a doubling time of less than a year (Fig. 8).

61 12 10 Delivered Cost of Solar , Active Space Heat X I CO a: to O a. tfl tr. O O Saving from Solar Delivered Cost of Heat from Distillate Fuel Oil I 1980 1990 2000 YEAR 2010 Figure 6 Estimated cost impact of solar technology on residential space heating

62 BY SECTOR I i Domestic, Farm Commercial Er id- Use Gro ss Industry Primary Transport i i Losses BY TYPE Heat Fluids <100°C 100°-140°C 140°-260° C >260 C Necessary Electricity Transport - Feedstocks - Losses (a) Fossil Fuels BY SOURCE Hydroelectric Coal Natural Gas Oil and Liquefied Petroleum Gas (Wood and Nuclear 1% Each) "Super Technical Fix" "Soft Sources" ! i Wind " J Feedstocks Transport Fluids 1 Liqwood (fuel alcohols from forestry) Hydroelectric Electricity Wood — _ Biogas ' i He >260°C at 140° -260° C Low. Temperature Solar Heat 1 <140°C (b) Figure 7 Canadian energy use: (a) in 1973 (total 6.6 quads, population 22 million) and (b) in 2025 (total 5.9 quads, population 40 million), as projected by assuming (left) "super technical fix" and (right) "soft" sources (Adapted from Lovins, 1976)

63 1,000r ki_ § LL O O _l I 100 10 Doubling Time 7.8 months 1971 1973 FISCAL YEAR 1975 1977 Figure 8 Federal budget authority for solar energy research and development funded by the National Science Foundation up to 1975 and the Energy Research and Development Administration from 1975 to 1977 (Solar Resource Group, 1976) Yet, for what needs to be done, this budget is small, and its em- phasis on solar electricity is almost perverse. On the face of it, a national decision to "go solar" seems unlikely. There is, however, at least one precedent from the recent past in which the direction of government action was decisively reoriented. The environmental legislation of the late 1960's and early 1970's was enacted through pressure brought by a coalition of interest groups having diverse beliefs. Some of the factors that could motivate the formation of a coalition capable of rapidly shifting governmental emphasis to re- newable forms of energy are the following: • Climatic Modifications. At present, there is growing evidence for the view that release of carbon dioxide or particulates into the atmosphere will have a long-term impact on global

64 climate. As analytic techniques improve, a consensus may develop among scientists concerning the likelihood of major climatic modifications if fossil fuels continue to be used. Such a consensus could lead to decisions to deemphasize their use. • Nuclear Accidents, Diversion, Waste Management. Major nu- clear accidents in the United States or elsewhere could lead to international reaction against any large-scale im- plementation of nuclear power. • Environmental Sensitivity. Within the past few years many laws have been passed in the United States and elsewhere en- couraging preservation of the environment. Our assessment of the energy potential of solar technologies is optimistic but credible. Because of the constraints on the development of alternative energy technologies discussed above, public speculation on the long-term social implications of solar energy remains, for the most part, incomplete. Solar-powered devices and techniques—in part because they are still novel—attract large audiences. But then, the adoption of new ideas and products has always been part of the American tradition. Witness the success of electronic calculators, which made mechanical calculators obsolete almost overnight. We assume that the American people and the government officials representing them will respond pragmatically to the present challenge. The combination of public opinion and shifting priorities on renewable forms of energy in government energy policies could stimulate action in the following ways: the sharp rise in energy prices could encour- age investment in solar energy earlier than might be expected on econ- omic grounds; the specter of energy shortages could encourage a rapid transition to renewable resources; governmental incentives, such as tax deductions, could encourage renewable technologies; and changes in values that emphasize scarcity, thrift, simplicity, or survival could discourage continuing use of conventional energy forms, despite their advantageous costs, for purposes such as generating electricity.

65 REFERENCES Demand and Conservation Panel. 1976. Draft report to the Committee on Nuclear and Alternative Energy Systems, National Research Council, Washington, B.C. December. Dow Chemical Co. 1975. Evaluation of New Energy Sources for Process Heat. Prepared for the National Science Foundation. Midland, Mich.: Dow Chemical Co. Haefele, W., and W. Sassin. 1976. Contrasting Views of the Future and Their Influence on our Technological Horizons for Energy. In Future Strategies for Energy Development: A Question of Scale. Oak Ridge, Tenn.: Oak Ridge Associated Universities. Hubbert, M. K. 1974. U.S. Energy Resources: A Review as of 1972, Pt. 1. In A National Fuels and Energy Policy Study. U.S. Congress, Senate, Committee on Interior and Insular Affairs. Serial No. 93-40 (92-75). Washington, D.C.: U.S. Government Printing Office. Lovins, A. 1976. Scale, Centralization and Electrification in Energy Systems. In Future Strategies for Energy Development: A Question of Scale. Oak Ridge, Tenn.: Oak Ridge Associated Universities. Marchetti, C. 1975. Geoengineering and the Energy Island. Second Status Report of the IIASA Project on Energy Systems. Laxenberg, Austria: International Institute for Applied Systems Analysis (RR-76-1). National Research Council. 1978. Energy Modeling for an Uncertain Future. Modeling Resource Group, Synthesis Panel, Committee on Nuclear and Alternative Energy Systems. Washington, D.C.: National Academy of Sciences. National Research Council. 1979. Domestic Potential of Solar and Other Renewable Energy Sources. Solar Resource Group, Supply and Delivery Panel, Committee on Nuclear and Alternative Energy Systems. Washington, D.C.: National Academy of Sciences. Schoen, R., A. S. Hirschberg, J. M. Weingart, and J. Stein. 1975. New Energy Technology for Buildings. Cambridge, Mass.: Ballinger. Solar Resource Group. 1976. Draft report to the Supply and Delivery Panel, Committee on Nuclear and Alternative Energy Systems, National Research Council, Washington, D.C. August 9. Stanford Research Institute. 1976. A Preliminary Social and Environmental Assessment of the ERDA Solar Program 1975-2020. Draft D Final Report, July. Menlo Park, Calif.: Stanford Research Institute.

66 Unger, Samuel G. 1975. Energy Utilization in the Leading Energy- Consuming Food Processing Industries. Food Technology 29(December): 33-46. U.S. Energy Research and Development Administration. 1975. A National Plan for Energy Research, Development and Demonstration. Washington, D.C.: U.S. Government Printing Office (ERDA-48). U.S. Energy Research and Development Administration. 1976. A National Plan for Energy Research, Development and Demonstration: Creating Energy Choices for the Future. Washington, D.C.: U.S. Government Printing Office (ERDA-76-1). Workshop on Alternative Energy Strategies. 1976. Energy Demand Studies: Major Consuming Countries. Cambridge, Mass.: MIT Press.