Renewable Energy:
Economic and Environmental Issues

by David Pimentel, G. Rodrigues, T. Wane, R. Abrams, K. Goldberg, H. Staecker, E. Ma, L. Brueckner, L. Trovato, C. Chow, U. Govindarajulu, and S. Boerke

(Originally published in BioScience -- Vol. 44, No. 8, September 1994)

Solar energy technologies, paired with energy conservation, have the potential to meet a large portion of future US energy needs

The United States faces serious energy shortages in the near future. High energy consumption and the ever-increasing US population will force residents to confront the critical problem of dwindling domestic fossil energy supplies. With only 4.7% of the world's population, the United States consumes approximately 25% of the total fossil fuel used each year throughout the world. The United States now imports about one half of its oil (25% of total fossil fuel) at an annual cost of approximately $65 billion (USBC 1992a). Current US dependence on foreign oil has important economic costs (Gibbons and Blair 1991) and portends future negative effects on national security and the economy.

Domestic fossil fuel reserves are being rapidly depleted, and it would be a major drain on the economy to import 100% of US oil. Within a decade or two US residents will be forced to turn to renewable energy for some of their energy needs. Proven US oil reserves are projected to be exhausted in 10 to 15 years depending on consumption patterns (DOE 1991a, Matare 1989, Pimentel et al. 1994, Worldwatch Institute 1992), and natural gas reserves are expected to last slightly longer. In contrast, coal reserves have been projected to last approximately 100 years, based on current use and available extraction processes (Matare 1989).

The US coal supply, however, could be used up in a much shorter period than the projected 100 years, if one takes into account predicted oil and gas depletion and concurrent population growth (DOE 1991a, Matare 1989). The US population is projected to double to more than one-half billion within the next 60 years (USBC 19921). How rapidly the coal supply is depleted will depend on energy consumption rates. The rapid depletion of US oil and gas reserves is expected to necessitate increased use of coal. By the year 2010, coal may constitute as much as 40% of total energy use (DOE 1991a). Undoubtedly new technologies will be developed that will make it possible to extract more oil and coal. However, this extra extraction can only be achieved at greater energy and economic costs. When the energy input needed to power these methods approaches the amount of energy mined, extraction will no longer be energy cost-effective (Hall et al. 1986).

Fossil fuel combustion, especially that based on oil and coal, is the major contributor to increasing carbon dioxide concentration in the atmosphere, thereby contributing to probable global warming. This climate change is considered one of the most serious environmental threats throughout the world because of its potential impact on food production and processes vital to a productive environment. Therefore, concerns about carbon dioxide emissions may discourage widespread dependence on coal use and encourage the development and use of renewable energy technologies.

Even if the rate of increase of per capita fossil energy consumption is slowed by conservation measures, rapid population growth is expected to speed fossil energy depletion and intensify global warming. Therefore, the projected availability of all fossil energy reserves probably has been overstated. Substantially reducing US use of fossil fuels through the efficient use of energy and the adoption of solar energy technologies extends the life of fossil fuel resources and could provide the time needed to develop and improve renewable energy technologies.

Renewable energy technologies will introduce new conflicts. For example, a basic parameter controlling renewable energy supplies is the availability of land. At present more than 99% of the US and world food supply comes from the land (FAO 1991). In addition, the harvest of forest resources is presently insufficient to meet US needs and thus the United States imports some of its forest products (USBC 1992a). With approximately 75% of the total US land area exploited for agriculture and forestry, there is relatively little land available for other uses, such as biomass production and solar technologies. Population growth is expected to further exacerbate the demands for land. Therefore, future land conflicts could be intense.

In this article, we analyze the potential of various renewable or solar energy technologies to supply the United States with its future energy needs. Diverse renewable technologies are assessed in terms of their land requirements, environmental benefits and risks, economic costs, and a comparison of their advantages. In addition, we make a projection of the amount of energy that could be supplied by solar energy subject to the constraints of maintaining the food and forest production required by society. Although renewable energy technologies often cause fewer environmental problems than fossil energy systems, they require large amounts of land and therefore compete with agriculture, forestry, and other essential land-use systems in the United States.

Assessment of renewable energy technologies

Coal, oil, gas, nuclear, and other mined fuels currently provide most of US energy needs. Renewable energy technologies provide only 8% (Table 1).

The use of solar energy is, however, expected to grow. Renewable energy technologies that have the potential to provide future energy supplies include: biomass systems, hydroelectric systems, hydrogen fuel, wind power, photovoltaics, solar thermal systems, and passive and active heating and cooling systems.

Biomass energy systems

At present, forest biomass energy, harvested from natural forests, provides an estimated 3.6 quads (1.1 x 10 18 Joules) or 4.2% of the US energy supply (Table 1). Worldwide, and especially in developing countries, biomass energy is more widely used than in the United States. Only forest biomass will be included in this US assessment, because forest is the most abundant biomass resource and the most concentrated form of biomass. However, some biomass proponents are suggesting the use of grasses, which on productive soils can yield an average of 5 t · ha-1 yr-1 (Hall et al. 1993, USDA 1992).

Although in the future most biomass probably will be used for space and water heating, we have analyzed its conversion into electricity in order to clarify the comparison with other renewable technologies. An average of 3 tons of (dry) woody biomass can be sustainably harvested per hectare per year with small amounts of nutrient fertilizer inputs (Birdsey 1992). This amount of woody biomass has a gross energy yield of 13.5 million kcal (thermal). The net yield is, however, lower because approximately 33 liters of diesel fuel oil per hectare is expended for cutting and collecting wood and for transportation, assuming an 80 kilometer roundtrip between the forest and the plant. The economic benefits of biomass are maximized when biomass can be used close to where it is harvested.

A city of 100,000 people using the biomass from a sustainable forest (3 tons/ha) for fuel would require approximately 220,000 ha of forest area, based on an electrical demand of 1 billion kWh (860 x 109 kcal = 1 kWh) per year (Table 2). Nearly 70% of the heat energy produced from burning biomass is lost in the conversion into electricity, similar to losses experienced in coal fired plants. The area required is about the same as that currently used by 100,000 people for food production, housing, industry, and roadways (USDA 1992).

The energy input/output ratio of this system is calculated to be 1:3 (Table 2). The cost of producing a kilowatt of electricity from woody biomass ranges from 7˘ to 10˘ (Table 2), which is competitive for electricity production that presently has a cost ranging from 3˘ to 13˘ (Table 2; USBC 1992a). Approximately 3 kcal of thermal energy is required to produce 1 kcal of electricity. Biomass could supply the nation with 5 quads of its total gross energy supply by the year 2050 with the use of at least 75 million ha (an area larger than Texas, or approximately 8% of the 917 million ha in the United States) (Table 3).

However, several factors limit reliance on woody biomass. Certainly, culturing fast-growing trees in a plantation system located on prime land might increase yields of woody biomass. However, this practice is unrealistic because prime land is essential for food production. Furthermore, such intensely managed systems require additional fossil fuel inputs for heavy machinery, fertilizers, and pesticides, thereby diminishing the net energy available. In addition, Hall et al. (1986) point out that energy is not the highest priority use of trees.

If natural forests are managed for maximal biomass energy production, loss of biodiversity can be expected. Also, the conversion of natural forests into plantations increases soil erosion and water runoff. Continuous soil erosion and degradation would ultimately reduce the overall productivity of the land. Despite serious limitations of plantations, biomass production could be increased using agroforestry technologies designed to protect soil quality and conserve biodiversity. In these systems, the energy and economic costs would be significant and therefore might limit the use of this strategy.

The burning of biomass is environmentally more polluting than gas but less polluting than coal. Biomass combustion releases more than 100 different chemical pollutants into the atmosphere (Alfheim and Ramdahl 1986). Wood smoke is reported to contain pollutants known to cause bronchitis, emphysema, and other illnesses. These pollutants include up to 14 carcinogens, 4 cocarcinogens, 6 toxins that damage cilia, and additional mucus-coagulating agents (Alfheim and Ramdahl 1986, DOE 1980). Of special concern are the relatively high concentrations of potentially carcinogenic polycyclic aromatic hydrocarbons (PAHs, organic compounds such as benzo(a)pyrene) and particulates found in biomass smoke (DOE 1980). Sulfur and nitrogen oxides, carbon monoxide, and aldehydes also are released in small though significant quantities and contribute to reduced air quality (DOE 1980). In electric generating plants, however, as much as 70% of these air pollutants can be removed by installing the appropriate air-pollution control devices in the combustion system.

Because of pollutants, several communities (including Aspen, Colorado) have banned the burning of wood for heating homes. When biomass is burned continuously in the home for heating, its pollutants can be a threat to human health (Lipfert et al. 1988, Smith 1987b).

When biomass in the form of harvested crop residues is used for fuel, the soil is exposed to intense erosion by wind and water (Pimentel et al. 1984). In addition to the serious degradation of valuable agricultural land, the practice of burning crop residues as a fuel removes essential nutrients from the land and requires the application of costly fossil-based fertilizers if yields are to be maintained. However, the soil organic matter, soil biota, and water-holding capacity of the soil cannot be replaced by applying fertilizers. Therefore, we conclude that crop residues should not be removed from the land for a fuel source (Pimentel 1992).

Biomass will continue to be a valuable renewable energy resource in the future, but its expansion will be greatly limited. Its use conflicts with the needs of agricultural and forestry production and contributes to major environmental problems.

Liquid fuels

Liquid fuels are indispensable to the US economy (DOE 1991a). Petroleum, essential for the transportation sector as well as the chemical industry, makes up approximately 42% of total US energy consumption. At present, the United States imports about one half of its petroleum and is projected to import nearly 100% within 10 to 15 years (DOE 1991a). Barring radically improved electric battery technologies, a shift from petroleum to alternative liquid and gaseous fuels will have to be made. The analysis in this section is focused on the potential of three liquid fuels: ethanol, methanol, and hydrogen.

Ethanol. A wide variety of starch and sugar crops, food processing wastes, and woody materials (Lynd et al. 1991) have been evaluated as raw materials for ethanol production. In the United States, corn appears to be the most feasible biomass feedstock in terms of availability and technology (Pimentel 1991).

The total fossil energy expended to produce 1 liter of ethanol from corn is 10,200 kcal, but note that 1 liter of ethanol has an energy value of only 5130 kcal. Thus, there is an energy imbalance causing a net energy loss. Approximately 53% of the total cost (55˘ per liter) of producing ethanol in a large, modern plant is for the corn raw material (Pimentel 1991). The total energy inputs for producing ethanol using corn can be partially offset when the dried distillers grain produced is fed to livestock. Although the feed value of the dried distillers grain reduces the total energy inputs by 8 % to 24%, the energy budget remains negative.

The major energy input in ethanol production, approximately 40% overall, is fuel needed to run the distillation process (Pimentel 1991). This fossil energy input contributes to a negative energy balance and atmospheric pollution. In the production process, special membranes can separate the ethanol from the so-called beer produced by fermentation. The most promising systems rely on distillation to bring the ethanol concentration up to 90%, and selective-membrane processes are used to further raise the ethanol concentration to 99.5% (Maeda and Kai 1991). The energy input for this upgrading is approximately 1280 kcal/liter. In laboratory tests, the total input for producing a liter of ethanol can potentially be reduced from 10,200 to 6200 kcal by using membranes, but even then the energy balance remains negative.

Any benefits from ethanol production, including the corn by-products, are negated by the environmental pollution costs incurred from ethanol production (Pimentel 1991). Intensive corn production in the United States causes serious soil erosion and also requires the further draw-down of groundwater resources. Another environmental problem is caused by the large quantity of stillage or effluent produced. During the fermentation process approximately 13 liters of sewage effluent is produced and placed in the sewage system for each liter of ethanol produced.

Although ethanol has been advertised as reducing air pollution when mixed with gasoline or burned as the only fuel, there is no reduction when the entire production system is considered. Ethanol does release less carbon monoxide and sulfur oxides than gasoline and diesel fuels. However, nitrogen oxides, formaldehydes, other aldehydes, and alcohol--all serious air pollutants-- are associated with the burning of ethanol as fuel mixture with or without gasoline (Sillman and Samson 1990). Also, the production and use of ethanol fuel contribute to the increase in atmospheric carbon dioxide and to global warming, because twice as much fossil energy is burned in ethanol production than is produced as ethanol.

Ethanol produced from corn clearly is not a renewable energy source. Its production adds to the depletion of agricultural resources and raises ethical questions at a time when food supplies must increase to meet the basic needs of the rapidly growing world population.

Methanol. Methanol is another potential fuel for internal combustion engines (Kohl 1990). Various raw materials can be used for methanol production, including natural gas, coal, wood, and municipal solid wastes. At present, the primary source of methanol is natural gas. The major limitation in using biomass for methanol production is the enormous quantities needed for a plant with suitable economies of scale. A suitably large methanol plant would require at least 1250 tons of dry biomass per day for processing (ACTI 1983). More than 150,000 ha of forest would be needed to supply one plant. Biomass generally is not available in such enormous quantities from extensive forests and at acceptable prices (ACTI 1983).

If methanol from biomass (33 quads) were used as a substitute for oil in the United States, from 250 to 430 million ha of land would be needed to supply the raw material. This land area is greater than the 162 million ha of US cropland now in production (USDA 1992). Although methanol production from biomass may be impractical because of the enormous size of the conversion plants (Kohl 1990), it is significantly more efficient than the ethanol production system based on both energy output and economics (Kohl 1990).

Compared to gasoline and diesel fuel, both methanol and ethanol reduce the amount of carbon monoxide and sulfur oxide pollutants produced, however both contribute other major air pollutants such as aldehydes and alcohol. Air pollutants from these fuels worsen the tropospheric ozone problem because of the emissions of nitrogen oxides from the richer mixtures used in the combustion engines (Sillman and Samson 1990).

Hydrogen. Gaseous hydrogen, produced by the electrolysis of water, is another alternative to petroleum fuels. Using solar electric technologies for its production, hydrogen has the potential to serve as a renewable gaseous and liquid fuel for transportation vehicles. In addition hydrogen can be used as an energy storage system for electrical solar energy technologies, like photovoltaics (Winter and Nitsch 1988).

The material inputs for a hydrogen production facility are primarily those needed to build a solar electric production facility. The energy required to produce 1 billion kWh of hydrogen is 1.3 billion kWh of electricity (Voigt 1984). If current photovoltaics (Table 2) require 2700 ha/1 billion kWh, then a total area of 3510 ha would be needed to supply the equivalent of 1 billion kWh of hydrogen fuel. Based on US per capita liquid fuel needs, a facility covering approximately 0.15 ha (16,300 ft2) would be needed to produce a year's requirement of liquid hydrogen. In such a facility, the water requirement for electrolytic production of 1 billion kWh/yr equivalent of hydrogen is approximately 300 million liters/yr (Voigt 1984).

To consider hydrogen as a substitute for gasoline: 9.5 kg of hydrogen produces energy equivalent to that produced by 25 kg of gasoline. Storing 25 kg of gasoline requires a tank with a mass of 17 kg, whereas the storage of 9.5 kg of hydrogen requires 55 kg (Peschka 1987). Part of the reason for this difference is that the volume of hydrogen fuel is about four times greater than that for the same energy content of gasoline. Although the hydrogen storage vessel is large, hydrogen burns 1.33 times more efficiently than gasoline in automobiles (Bockris and Wass 1988). In tests, a BMW 745i liquid hydrogen test vehicle with a tank weight of 75 kg, and the energy equivalent of 40 liters (320,000 kcal) of gasoline, had a cruising range in traffic of 400 km or a fuel efficiency of 10 km per liter (24 mpg) (Winter 1986).

At present, commercial hydrogen is more expensive than gasoline. For example, assuming 5˘ per kWh of electricity from a conventional power plant, hydrogen would cost 9˘ per kWh (Bockris and Wass 1988). This cost is the equivalent of 67˘/liter of gasoline. Gasoline sells at the pump in the United States for approximately 30˘/liter. However, estimates are that the real cost of burning a liter of gasoline ranges from $1.06 to $1.32, when production, pollution, and other external costs are included (Worldwatch Institute 1989). Therefore, based on these calculations hydrogen fuel may eventually be competitive.

Some of the oxygen gas produced during the electrolysis of water can be used to offset the cost of hydrogen. Also the oxygen can be combined with hydrogen in a fuel cell, like those used in the manned space flights. Hydrogen fuel cells used in rural and suburban areas as electricity sources could help decentralize the power grid, allowing central power facilities to decrease output, save transmission costs, and make mass-produced, economical energy available to industry.

Compared with ethanol, less land (0.15 ha versus 7 ha for ethanol) is required for hydrogen production that uses photovoltaics to produce the needed electricity. The environmental impacts of hydrogen are minimal. The negative impacts that occur during production are all associated with the solar electric technology used in production.

Water for the production of hydrogen may be a problem in the arid regions of the United States, but the amount required is relatively small compared with the demand for irrigation water in agriculture. Although hydrogen fuel produces emissions of nitrogen oxides and hydrogen peroxide pollutants, the amounts are about one-third lower than those produced from gasoline engines (Veziroglu and Barbir 1992). Based on this comparative analysis, hydrogen fuel may be a cost-effective alternative to gasoline, especially if the environmental and subsidy costs of gasoline are taken into account.

Hydroelectric systems

For centuries, water has been used to provide power for various systems. Today hydropower is widely used to produce electrical energy. In 1988 approximately 870 billion kWh (3 quads or 9.5 % ) of the United States' electrical energy was produced by hydroelectric plants (FERC 1988, USBC 1992a). Further development and/or rehabilitation of existing dams could produce an additional 48 billion kWh per year. However, most of the best candidate sites already have been fully developed, although some specialists project increasing US hydropower by as much as 100 billion kWh if additional sites are developed (USBC 1992a).

Hydroelectric plants require land for their water-storage reservoirs. An analysis of 50 hydroelectric sites in the United States indicated that an average of 75,000 ha of reservoir area are required per 1 billion kWh/ yr produced (Table 2). However, the size of reservoir per unit of electricity produced varies widely, ranging from 482 ha to 763,000 ha per 1 billion kWh/yr depending upon the hydro head, terrain, and additional uses made of the reservoir (Table 2). The latter include flood control, storage of water for public and irrigation supplies, and/or recreation (FERC 1984). For the United States the energy input/output ratio was calculated to be 1:48 (Table 2); for Europe an estimate of 1:15 has been reported (Winter et al. 1992).

Based on regional estimates of land use and average annual energy generation, approximately 63 million hectares of the total of 917 million ha of land area in the United States are currently covered with reservoirs. To develop the remaining best candidate sites, assuming land requirements similar to those in past developments, an additional 24 million hectares of land would be needed for water storage (Table 3).

Reservoirs constructed for hydroelectric plants have the potential to cause major environmental problems. First, the impounded water frequently covers agriculturally productive, alluvial bottomland. This water cover represents a major loss of productive agricultural land. Dams may fail, resulting in loss of life and destruction of property. Further, dams alter the existing plant and animal species in the ecosystem (Flavin 1985). For example, cold water fishes may be replaced by warm water fishes, frequently blocking fish migration (Hall et al. 1986). However, flow schedules can be altered to ameliorate many of these impacts. Within the reservoirs, fluctuations of water levels alter shorelines and cause downstream erosion and changes in physiochemical factors, as well as the changes in aquatic communities. Beyond the reservoirs, discharge patterns may adversely reduce downstream water quality and biota, displace people, and increase water evaporation losses (Barber 1993). Because of widespread public environmental concerns, there appears to be little potential for greatly expanding either large or small hydroelectric power plants in the future (Table 3).

Wind power

For many centuries, wind power like water power has provided energy to pump water and run mills and other machines. In rural America windmills have been used to generate electricity since the early 1900s.

Modern wind turbine technology has made significant advances over the last 10 years. Today, small wind machines with 5 to 40 kW capacity can supply the normal electrical needs of homes and small industries (Twidell 1987). Medium-size turbines rated 100 kW to 500 kW produce most of the commercially generated electricity. At present, the larger, heavier blades required by large turbines upset the desirable ratio between size and weight and create efficiency problems. However, the effectiveness and efficiency of the large wind machines are expected to be improved through additional research and development of lighter weight but stronger components (Clarke 1991). Assuming a 35% operation capacity at a favorable site, the energy input/output ratio of the system is 1:5 for the material used in the construction of medium size wind machines (Table 2).

The availability of sites with sufficient wind (at least 20 km/in) limits the widespread development of wind farms. Currently, 70% of the total wind energy (0.01 quad) produced in the United States is generated in California (Table 3; AWEA 1992). However, an estimated 13% of the contiguous US land area has wind speeds of 22 km/in or higher; this area then would be sufficient to generate approximately 20 % of US electricity using current technology (DOE 1992). Promising areas for wind development include the Great Plains and coastal regions.

Another limitation of this energy resource is the number of wind machines that a site can accommodate. For example, at Altamont Pass, California, an average of one turbine per 1.8 ha allows sufficient spacing to produce maximum power (Smith and Ilyin 1991). Based on this figure approximately 11,700 ha of land are needed to supply 1 billion kWh/ yr (Table 2). However, because the turbines themselves only occupy approximately 2% of the area or 230 ha, dual land use is possible. For example, current agricultural land developed for wind power continues to be used in cattle, vegetable, and nursery stock production.

An investigation of the environmental impacts of wind energy production reveals a few hazards. For example, locating the wind turbines in or near the flyways of migrating birds and wildlife refuges may result in birds flying into the supporting structures and rotating blades (Clarke 1991, Kellett 1990). Clarke suggests that wind farms be located at least 300 meters from nature reserves to reduce this risk to birds.

Insects striking turbine blades will probably have only a minor impact on insect populations, except for some endangered species. However, significant insect accumulation on the blades may reduce turbine efficiency (Smith 1987a).

Wind turbines create interference with electromagnetic transmission, and blade noise may be heard up to 1 km away (Kellet 1990). Fortunately, noise and interference with radio and television signals can be eliminated by appropriate blade materials and careful placement of turbines. In addition, blade noise is offset by locating a buffer zone between the turbines and human settlements. New technologies and designs may minimize turbine generator noise.

Under certain circumstances shadow flicker has caused irritation, disorientation, and seizures in humans (Steele 1991). However, as with other environmental impacts, mitigation is usually possible through careful site selection away from homes and offices. This problem slightly limits the land area suitable for wind farms.

Although only a few wind farms supply power to utilities in the United States, future widespread development may be constrained because local people feel that wind farms diminish the aesthetics of the area (Smith 1987a). Some communities have even passed legislation to prevent wind turbines from being installed in residential areas (Village of Cayuga Heights, New York, Ordinance 1989). Likewise areas used for recreational purposes, such as parks, limit the land available for wind power development.

Photovoltaics

Photovoltaic cells are likely to provide the nation with a significant portion of its future electrical energy (DeMeo et al 1991). Photovoltaic cells produce electricity when sunlight excites electrons in the cells. Because the size of the units is flexible and adaptable, photovoltaic cells are ideal for use in homes, industries, and utilities.

Before widespread use, however, improvements are needed in the photovoltaic cells to make them economically competitive. Test photovoltaic cells that consist of silicon solar cells are currently up to 21% efficient in converting sunlight into electricity (Moore 1992). The durability of photovoltaic cells, which is now approximately 20 years, needs to be lengthened and current production costs reduced about fivefold to make them economically feasible. With a major research investment, all of these goals appear possible to achieve (DeMeo et al. 1991).

Currently, production of electricity from photovoltaic cells costs approximately 30˘/kWh, but the price is projected to fall to approximately 10˘/kWh by the end of the decade and perhaps reach as low as 4˘ by the year 2030, provided the needed improvements are made (Flavin and Lenssen 1991). In order to make photovoltaic cells truly competitive, the target cost for modules would have to be approximately 8˘/ kWh (DeMeo et al. 1991).

Using photovoltaic modules with an assumed 7.3% efficiency (the current level of commercial units), 1 billion kWh/yr of electricity could be produced on approximately 2700 ha of land (Table 2), or approximately 0.027 ha per person, based on the present average per capita use of electricity. Thus, total US electrical needs theoretically could be met with photovoltaic cells on 5.4 million ha (0.6% of US land). If 21% efficient cells were used, the total area needed would be greatly reduced. Photovoltaic plants with this level of efficiency are being developed (DeMeo et al. 1991).

The energy input for the structural materials of a photovoltaic system delivering 1 billion kWh is calculated to be approximately 300 kWh/m2. The energy input/output ratio for production is about 1:9 assuming a life of 20 years (Table 2).

Locating the photovoltaic cells on the roofs of homes, industries, and other buildings would reduce the need for additional land by approximately 5% (USBC 1992a), as well as reduce the costs of energy transmission. However, photovoltaic systems require back­up with conventional electrical systems, because they function only during daylight hours.

Photovoltaic technology offers several environmental advantages in producing electricity compared with fossil fuel technologies. For example, using present photovoltaic technology, carbon dioxide emissions and other pollutants are negligible.

The major environmental problem associated with photovoltaic systems is the use of toxic chemicals such as cadmium sulfide and gallium arsenide, in their manufacture (Holdren et al. 1980). Because these chemicals are highly toxic and persist in the environment for centuries, disposal of inoperative cells could become a major environmental problem. However, the most promising cells in terms of low cost, mass production, and relatively high efficiency are those being manufactured using silicon. This material makes the cells less expensive and environmentally safer than the heavy metal cells.

Solar thermal conversion systems

Solar thermal energy systems collect the sun's radiant energy and convert it into heat. This heat can be used for household and industrial purposes and also to drive a turbine and produce electricity. System complexity ranges from solar ponds to the electric-generating central receivers. We have chosen to analyze electricity in order to facilitate comparison to the other solar energy technologies.

Solar ponds. Solar ponds are used to capture solar radiation and store it at temperatures of nearly 100°C. Natural or man-made ponds can be made into solar ponds by creating a salt-concentration gradient made up of layers of increasing concentrations of salt. These layers prevent natural convection from occurring in the pond and enable heat collected from solar radiation to be trapped in the bottom brine.

The hot brine from the bottom of the pond is piped out for generating electricity. The steam from the hot brine turns freon into a pressurized vapor, which drives a Rankine engine. This engine was designed specifically for converting low-grade heat into electricity. At present, solar ponds are being used in Israel to generate electricity (Tabor and Doran 1990).

For successful operation, the salt concentration gradient and the water levels must be maintained. For example, 4000 ha of solar ponds lose approximately 3 billion liters of water per year under the arid conditions of the southwestern United States (Tabor and Doran 1990). In addition, to counteract the water loss and the upward diffusion process of salt in the ponds, the dilute salt water at the surface of the ponds has to be replaced with fresh water. Likewise salt has to be added periodically to the heat-storage zone. Evaporation ponds concentrate the brine, which can then be used for salt replacement in the solar ponds.

Approximately 4000 ha of solar ponds (40 ponds of 100 ha) and a set of evaporation ponds that cover a combined 1200 ha are needed for the production of 1 billion kWh of electricity needed by 100,000 people in one year (Table 2). Therefore, a family of three would require approximately 0.2 ha (22,000 sq ft) of solar ponds for its electricity needs. Although the required land area is relatively large, solar ponds have the capacity to store heat energy for days, thus eliminating the need for back-up energy sources from conventional fossil plants. The efficiency of solar ponds in converting solar radiation into heat is estimated to be approximately 1:5. Assuming a 30-year life for a solar pond, the energy input/output ratio is calculated to be 1:4 (Table 2). A 100 hectare (1 km2) solar pond is calculated to produce electricity at a rate of approximately 14˘ per kWh. According to Folchitto (1991), this cost could be reduced in the future.

In several locations in the United States solar ponds are now being used successfully to generate heat directly. The heat energy from the pond can be used to produce processed steam for heating at a cost of only 2˘ to 3.5˘ per kWh (Gommend and Grossman 1988). Solar ponds are most effectively employed in the Southwest and Mid-west.

Some hazards are associated with solar ponds, but most can be prevented with careful management. For instance, it is essential to use plastic liners to make the ponds leakproof and thereby prevent contamination of the adjacent soil and groundwater with salt. Burrowing animals must be kept away from the ponds by buried screening (Dickson and Yates 1983). In addition, the ponds should be fenced to prevent people and other animals from coming in contact with them. Because some toxic chemicals are used to prevent algae growth on water surface and freon is used in the Rankine engine, methods will have to be devised for safely handling these chemicals (Dickson and Yates 1983).

Solar receiver systems. Other solar thermal technologies that concentrate solar radiation for large scale energy production include distributed and central receivers. Distributed receiver technologies use rows of parabolic troughs to focus sunlight on a central-pipe receiver that runs above the troughs. Pressurized water and other fluids are heated in the pipe and are used to generate steam to drive a turbogenerator for electricity production or provide industry with heat energy.

Central receiver plants use computer-controlled, sun-tracking mirrors, or heliostats, to collect and concentrate the sunlight and redirect it toward a receiver located atop a centrally placed tower. In the receiver, the solar energy is captured as heat energy by circulating fluids, such as water or molten salts, that are heated under pressure. These fluids either directly or indirectly generate steam, which is then driven through a conventional turbogenerator to yield electricity. The receiver system may also be designed to generate heat for industry.

Distributed receivers have entered the commercial market before central receivers, because central receivers are more expensive to operate. But, compared to distributed receivers, central receivers have the potential for greater efficiency in electricity production because they are able to achieve higher energy concentrations and higher turbine inlet temperatures (Winter 1991). Central receivers are used in this analysis.

The land requirements for the central receiver technology are approximately 1100 ha to produce 1 billion kWh/yr (Table 2), assuming peak efficiency, and favorable sunlight conditions like those in the western United States. Proposed systems offer four to six hours of heat storage and may be constructed to include a back-up alternate energy source. The energy input/output ratio is calculated to be 1:10 (Table 2). Solar thermal receivers are estimated to produce electricity at approximately 10˘ per kWh, but this cost is expected to be reduced somewhat in the future, making the technology more competitive (Vant-Hull 1992). New technical advances aimed at reducing costs and improving efficiency include designing stretched membrane heliostats, volumetric-air ceramic receivers, and improved overall system designs (Beninga et al. 1991).

Central receiver systems are being tested in Italy, France, Spain, Japan, and the United States (at the 10-megawatt Solar One pilot plant near Barstow, California; Skinrood and Skvarna 1986). Also, Luz's Solar Electric Generating System plants at Barstow use distributed receivers to generate almost 300 MW of commercial electricity (Jensen et al. 1989).

The potential environmental impacts of solar thermal receivers include: the accidental or emergency release of toxic chemicals used in the heat transfer system (Baechler and Lee 1991); bird collisions with a heliostat and incineration of both birds and insects if they fly into the high temperature portion of the beams; and--if one of the heliostats did not track properly but focused its high temperature beam on humans, other animals, or flammable materials--burns, retinal damage, and fires (Mihlmester et al. 1980). Flashes of light coming from the heliostats may pose hazards to air and ground traffic (Mihlmester et al. 1980).

Other potential environmental impacts include microclimate alteration, for example reduced temperature and changes in wind speed and evapotranspiration beneath the heliostats or collecting troughs. This alteration may cause shifts in various plant and animal populations. The albedo in solar­collecting fields may be increased from 30% to 56% in desert regions (Mihlmester et al. 1980). An area of 1100 ha is affected by a plant producing 1 billion kWh.

The environmental benefits of receiver systems are significant when compared to fossil fuel electrical generation. Receiver systems cause no problems of acid rain, air pollution, or global warming (Kennedy et al. 1991).

Passive heating and cooling of buildings

Approximately 23% (18.4 quads) of the fossil energy consumed yearly in the United States is used for space heating and cooling of buildings and for heating hot water (DOE 1991a). At present only 0.3 quads of energy are being saved by technologies that employ passive and active solar heating and cooling of buildings (Table 2). Tremendous potential exists for substantial energy savings by increased energy efficiency and by using solar technologies for buildings.

Both new and established homes can be fitted with solar heating and cooling systems. Installing passive solar systems into the design of a new home is generally cheaper than retrofitting an existing home. Including passive solar systems during new home construction usually adds less than 10% to construction costs (Howard and Szoke 1992); a 3-5% added first cost is typical.1 Based on the cost of construction and the amount of energy saved measured in terms of reduced heating costs, we estimate the cost of passive solar systems to be approximately 3˘ per kWh saved.

Improvements in passive solar technology are making it more effective and less expensive than in the past. In the area of window designs, for example, current research is focused on the development of superwindows with high insulating values and smart or electrochromic windows that can respond to electrical current, temperature, or incident sunlight to control the admission of light energy (Warner 1991). Use of transparent insulation materials makes window designs that transmit from 50% to 70% of incident solar energy while at the same time providing insulating values typical of 25 cm of fiber glass insulation (Chahroudi 1992). Such materials have a wide range of solar technology applications beyond windows, including house heating with transparent, insulated collector­storage walls and integrated storage collectors for domestic hot water (Wittwer et al. 1991).

Active solar heating technologies are not likely to play a major role in the heating of buildings. The cost of energy saved is relatively high compared with passive systems and conservation measures.2

Solar water heating is also cost effective. Approximately 3% of all the energy used in the United States is for heating water in homes (DOE 1991a). In addition, many different types of passive and active water heating solar systems are available and are in use throughout the United States. These systems are becoming increasingly affordable and reliable (Wittwer et al. 1991). The cost of purchasing and installing an active solar water heater ranges from $2500 to $6000 in the northern regions and $2000 to $4000 in the southern regions of the nation (DOE/ CE 1988).

Although none of the passive heating and cooling technologies require land, they can cause environmental problems. For example, some indirect land-use problems may occur, such as the removal of trees, shading, and rights to the sun (Schurr et al. 1979). Glare from collectors and glazing could create hazards to automobile drivers, pedestrians, bicyclists, and airline pilots. Also, when houses are designed to be extremely energy efficient and airtight, indoor air quality becomes a concern because air pollutants may accumulate inside. However, installation of well­designed ventilation systems promotes a healthful exchange of air while reducing heat loss during the winter and heat gain during the summer. If radon is a pollutant present at unsafe levels in the home, various technologies can mitigate the problem (ASTM 1992).

Comparing solar power to coal and nuclear power

Coal and nuclear power production are included in this analysis to compare conventional sources of electricity generation to various future solar energy technologies. Coal, oil, gas, nuclear, and other mined fuels are used to meet 92% of US energy needs (Table 1). Coal and nuclear plants combined produce three quarters of US electricity (USBC 1992a).

Energy efficiencies for both coal and nuclear fuels are low due to the thermal law constraint of electric generator designs: coal is approximately 35% efficient and nuclear fuels approximately 33% (West and Kreith 1988). Both coal and nuclear power plants in the future may require additional structural materials to meet clean air and safety standards. However, the energetic requirements of such modifications are estimated to be small compared with the energy lost due to conversion inefficiencies

The costs of producing electricity using coal and nuclear energy are 3 ˘ and 5˘ per kWh, respectively (EIA 1990). However, the costs of this kind of energy generation are artificially low because they do not include such external costs as damages from acid rain produced from coal and decommissioning costs for the closing of nuclear plants. The Clean Air Act and its amendments may raise coal generation costs, while the new reactor designs, standardization, and streamlined regulations may reduce nuclear generation costs. Government subsidies for nuclear and coal plants also skew the comparison with solar energy technologies (Wolfson 1991).

Clouding the economic costs of fossil energy use are the direct and indirect US subsidies that hide the true cost of energy and keep the costs low, thereby encouraging energy consumption. The energy industry receives a direct subsidy of $424 per household per year (based on an estimated maximum of $36 billion for total federal energy subsidies [ASK 1993]). In addition, the mined-energy industry, like the gasoline industry, does not pay for the environmental and public health costs of fossil energy production and consumption.

The land requirements for fossil fuel and nuclear-based plants are lower than those for solar energy technologies (Table 2). The land area required for electrical production of 1 billion kWh/year is estimated at 363 ha for coal and 48 ha for nuclear fuels. These figures include the area for the plants and both surface and underground mining operations and waste disposal. The land requirements for coal technology are low because it uses concentrated fuel sources rather than diffuse solar energy. However, as the quality of fuel ore declines, land requirements for mining will increase. In contrast, efficient reprocessing and the use of nuclear breeder reactors may decrease the land area necessary for nuclear power.

Many environmental problems are associated with both coal and nuclear power generation (Pimentel et al. 1994). For coal, the problems include the substantial damage to land by mining, air pollution, acid rain, global warming, as well as the safe disposal of large quantities of ash (Wolfson 1991). For nuclear power, the environmental hazards consist mainly of radioactive waste that may last for thousands of years, accidents, and the decommissioning of old nuclear plants (Wolfson 1991).

Fossil-fuel electric utilities account for two-thirds of the sulfur dioxide, one-third of the nitrogen dioxide, and one-third of the carbon dioxide emissions in the United States (Kennedy et al. 1991). Removal of carbon dioxide from coal plant emissions could raise costs to 12˘/kWh; a disposal tax on carbon could raise coal electricity costs to 18˘/kWh (Williams et al. 1990).

The occupational and public health risks of both coal and nuclear plants are fairly high, due mainly to the hazards of mining, ore transportation, and subsequent air pollution during the production of electricity. However, there are 22 times as many deaths per unit of energy related to coal than of nuclear energy production because 90,000 times greater volume of coal than nuclear ore is needed to generate an equivalent amount of electricity.3

Also, and as important, coal produces more diffuse pollutants than nuclear fuels during normal operation of the generating plant. Coal fired plants produce air pollutants-- including sulfur oxides, nitrogen oxides, carbon dioxide, and particulates--that adversely affect air quality and contribute to acid rain. Technologies do exist for removing most of the air pollutants, but their use increases the cost of a new plant by 20-25% (IEA 1987). By comparison, nuclear power produces many fewer pollutants than do coal plants (Tester et al. 1991).

Transition to solar energy and other alternatives

The first priority of a sustainable US energy program should be for individuals, communities, and industries to conserve fossil energy resources. Other developed countries have proven that high productivity and a high standard of living can be achieved with considerably less energy expenditure compared to that of the United States. Improved energy efficiency in the United States, other developed nations, and even in developing nations would help both extend the world's fossil energy resources and improve the environment (Pimentel et al. 1994).

The supply and demand for fossil and solar energy; the requirements of land for food, fiber, and lumber; and the rapidly growing human population will influence future US options. The growth rate of the US population has been increasing and is now at 1.1 % per year (USBC 1992b); at this rate, the present population of 260 million will increase to more than a half billion in just 60 years. The presence of more people will require more land for homes, businesses, and roads. Population density directly influences food production, forest product needs, and energy requirements. Considerably more agricultural and forest land will be needed to provide vital food and forest products, and the drain on all energy resources will increase. Although there is no cropland shortage at present (USDA 1992), problems undoubtedly will develop in the near future in response to the diverse needs of the growing US population.

Solar energy technologies, most of which require land for collection and production, will compete with agriculture and forestry in the United States and worldwide (Table 2). Therefore, the availability of land is projected to be a limiting factor in the development of solar energy. In the light of this constraint, an optimistic projection is that the current level of nearly 7 quads of solar energy collected and used annually in the United States could be increased to approximately 37 quads (Ogden and Williams 1989, Pimentel et al. 1984). This higher level represents only 43% of the 86 quads of total energy currently consumed in the United States (Tables 1 and 3). Producing 37 quads with solar technologies would require approximately 173 million ha, or nearly 20% of US land area (3). At present this amount of land is available, but it may become unavailable due to future population growth and increased resource consumption. If land continues to be available, the amounts of solar energy (including hydropower and wind) that could be produced by the year 2050 are projected to be: 5 quads from biomass, 4 quads from hydropower, 8 quads from wind power, 6 quads from solar thermal systems, 6 quads from passive and active solar heating, and 8 quads from photovoltaics (3).

Another possible future energy source is fusion energy (Bartlett 1994, Matare 1989). Fusion uses nuclear particles called neutrons to generate heat in a fusion reactor vessel. Nuclear fusion differs from fission in that the production of energy does not depend on continued mining. However, high costs and serious environmental problems are anticipated (Bartlett 1994). The environmental problems include the production of enormous amounts of heat and radioactive material.

The United States could achieve a secure energy future and a satisfactory standard of living for everyone if the human population were to stabilize at an estimated optimum of 200 million (down from today's 260 million) and conservation measures were to lower per capita energy consumption to about half the present level (Pimentel et al. 1994). However, if the US population doubles in 60 years as is more likely, supplies of energy, food, land, and water will become inadequate, and land, forest, and general environmental degradation will escalate (Pimentel et al. 1994, USBC 1992a).

Fossil energy subsidies should be greatly diminished or withdrawn and the savings should be invested to encourage the development and use of solar energy technologies. This policy would increase the rate of adoption of solar energy technologies and lead to a smooth transition from a fossil fuel economy to one based on solar energy. In addition, the nation that becomes a leader in the development of solar energy technologies is likely to capture the world market for this industry.

Conclusions

This assessment of alternate technologies confirms that solar energy alternatives to fossil fuels have the potential to meet a large portion of future US energy needs, provided that the United States is committed to the development and implementation of solar energy technologies and that energy conservation is practiced. The implementation of solar technologies will also reduce many of the current environmental problems associated with fossil fuel production and use.

An immediate priority IS to speed the transition from reliance on nonrenewable energy sources to reliance on renewable, especially solar based, energy technologies. Various combinations of solar technologies should be developed consistent with the characteristics of different geographic regions, taking into account the land and water available and regional energy needs. Combined, biomass energy and hydroelectric energy in the United States currently provide nearly 7 quads of solar energy, and their output could be increased to provide up to 9 quads by the year 2050. The remaining 28 quads of solar renewable energy needed by 2050 is projected to be produced by wind power, photovoltaics, solar thermal energy, and passive solar heating. These technologies should be able to provide energy without interfering with required food and forest production.

If the United States does not commit itself to the transition from fossil to renewable energy during the next decade or two, the economy and national security will be adversely affected. Starting immediately, it is paramount that US residents must work together to conserve energy, land, water, and biological resources. To ensure a reasonable standard of living in the future, there must be a fair balance between human population density and energy, land, water, and biological resources.

Acknowledgments

We thank the following people for reading an earlier draft of this article, for their many helpful suggestions, and in some cases, for providing additional information: A. Baldwin, Office of Technology Assessment, US Congress; A. A. Bartlett, University of Colorado, Boulder; E. DeMeo, Electric Power Research Institute; H. English, Passive Solar Industries Council; S. L. Frye, Bechtel; M. Giampietro, National Institute of Nutrition, Rome, Italy; J. Goldemberg, Universidade de Sao Paulo, Brazil; C. A. S. Hall, College of Environmental Science and Forestry, SUNY, Syracuse; D. O. Hall, King's College, London, United Kingdom; S. Harris, Oak Harbor, WA; J. Harvey, New York State Energy Research and Development Authority; B. D. Howard, The Alliance to Save Energy; C. V. Kidd, Washington, DC; N. Lenssen, Worldwatch Institute; L. R. Lynd, Dartmouth College; J. M. Nogueira, Universidade de Brasilia, Brazil; M. G. Paoletti, University of Padova, Italy; R. Ristenen, University of Colorado, Boulder; S. Sklar, Solar Energy Industries Association; R. Swisher, American Wind Energy Association; R. W. Tresher, National Renewable Energy Laboratory; L. L. Vant-Hull, University of Houston; Wang Zhaoqian, Zheijan Agricultural University, China; P. B. Weisz, University of Pennsylvania, Philadelphia; Wen Dazhong, Academia Sinica, China; C. J. Winter, Deutsche Forschungsanstalt fur Luft und Raumfahrt, Germany; D. L. Wise, Northeastern University, Boston; and at Cornell University: R. Barker, S. Bukkens, D. Hammer, L. Levitan, S. Linke, and M. Pimentel.

References cited

Advisory Committe on Technology Innovation (ACTI). 1983. Alcohol Fuels: Options for Developing Countries. National Academy Press, Washington, DC.

Alfheim, 1., and T. Ramdahl. 1986. Mutagenic and carcinogenic Compounds from Energy Generation Final Report No. NP-6752963 (NTIS No. DE 86752963). Center for Industriforkning. Oslo, Norway.

Alliance to Save Energy (ASE). 1993. Federal Energy Subsidies: Energy, Environmental, and Fiscal Impacts The Alliance to Save Energy, Washington, DC.

American Society for Testing Materials (ASTM). 1992. Standard guide for radon control options for the design and construction of new low rise residential buildings. Pages 1117-1123 in Annual/ Book of American Society for Testing Materials. E1465-92. ASTM, Philadelphia, PA.

American Wind Energy Association (AWEA). 1991. Wind energy comes of age. Solar Today 5: 14-16.

1992. Wind Technology Status Report. American Wind Energy Association, Washington, DC.

Baechler, M. C., and A. D. Lee. 1991. Implications of environmental externalities assessments for solar thermal power plants. Pages 151-158 in T. R. Mancini K. Watanabe and D. E. Klett, eds. Solar Engineering 1991. American Society of Mechanical Engineering, New York.

Barber, M. 1993. Why more energy? The hidden cost of Canada's cheap power. Forests, Trees and People Newsletter No. 19: 26-29.

Bartlett, A. A. 1994. Fusion: An illusion or a practical source of energy? Clearinghouse Bulletin 4(1): 1-3, 7.

Beninga, K., R. Davenport, J. Sandubrue, and K. Walcott. 1991. Design and fabrication of a market ready stretched membrane heliostat. Pages 229-234 in T. R. Mancini, K. Watanabe and D. E. Klett eds. Solar Engineering 1991. American Society of Mechanical Engineering, New York.

Birdsey, R. A. 1992. Carbon storage and accumulation in United States forest ecosystems. General Technical Report WO 59. USDA Forest Service, Washington, DC.

Bockris, J. O. M., and J. C. Wass. 1988. About the real economics of massive hydrogen production at 2010 AD. Pages 101-151 in T. N. Veziroglu and A. N. Protsenko, eds. Hydrogen Energy Vll. Pergamon Press, New York.

Chahroudi, D. 1992. Weather panel architecture: a passive solar solution for cloudy climates. Solar Today 6: 17-20.

Clarke, A. 1991. Wind energy progress and potential. Energy Policy 19: 742-755.

DeMeo, E. A., F. R. Goodman, T. M. Peterson, and J. C. Schaefer. 1991. Solar Photovoltaic Power: A US Electric Utility R & D Perspective. Edited by 2.1. P. S. conference. IEEE Photovoltaic Specialist Conference Proceedings, New York.

Dickson, Y. L., and B. C. Yates. 1983. Examination of the Environmental Effects of Salt-Gradient Solar Ponds in the Red River Basin of Texas. North Texas State University, Denton Institute of Applied Sciences, Denton, TX.

Department of Energy (DOE). 1980. Health Effects of Residential Wood Combustion: survey of Know ledge and Research. DOE, Technology Assessments Division, Washington, DC.

1990. The Potential of Renew able Energy. SERI, DOE, Golden, CO.

1991a. A Annual Energy Outlook with Projections to 2010. DOE, Energy Information Administration, Washington, DC.

199lb. 1989 International Energy Annual/. DOE, Washington, DC.

1992. Wind Energy Program Overview. National Renewable Energy Laboratory, DOE, Golden, CO.

DOE/Conservation Energy (CE). 1988. Passive and Active Solar Domestic Hot Water Systems. FS-119. DOE, Conservation and Renewable Energy Inquiry and Referral Service, Washington, DC.

Energy Information Administration (EIA). 1990. Electric Plant Cost and Power Production Expenses. EIA, Washington, DC.

Electric Power Research Institute (EPRI). 1991. Photovoltaic system performance assessment for 1989. EPRI Interim Report 65-7286. EPRI, Los Angeles, CA.

Food and Agriculture Organization (FAO). 1991. Food Balance Sheets. FAO, Rome.

Federal Energy Regulatory Commission (FERC). 1984. Hydroelectric Power Resources of the United States. FERC, Washington, DC.

1988. Hydroelectric Power Resources of the United States: Developed and Undeveloped. FERC, Washington, DC.

Flavin, C. 1985. Renewable Energy at the Crossroads. Center for Renewable Resources, Washington, DC.

Flavin, C., and N. Lenssen.1991. Here comes the sun. Pages 10-18 in L. Brown et al., ed s. State of the World 1991.Worldwatch Institute, Washington, DC.

Folchitto, S. 1991. Seawater as salt and water source for solar ponds. Sol. Energy 46: 343-51.

Gibbons, J. H., and P. D. Blair. 1991. US energy transition: on getting from here to there. Physics Today, 44: 22-30.

Gommend, K., and G. Grossman. 1988. Process steam generation by temperature boosting of heat from solar ponds. Solar Ponds 41: 81-89.

Hall, C. A. S., C. J. Cleveland, and R. L. Kaufmann. 1986. Energy and Resource Quality: The Ecology of the Economic Process. Wiley, New York.

Hall, D. O., F. Rosillo-Calle, R. H. Williams, and J. Woods. 1993. Biomass for energy: supply prospects. In T. B. Johansson, H. Kelley, A. K. N. Reddy, and R. H. Williams, eds. Renewable Energy. Island Press, Washington, DC.

Holdren, J. P., G. Morris, and 1. Mintzer. 1980. Environmental aspects of renewable energy. Annual Review of Energy 5: 241-291.

Howard, B., and S. S. Szoke. 1992. Advances in Solar Design Tools. 5th Thermal Envelope Conference, December 1992. US Dept. of Energy, ASHRAE, Clearwater, FL.

International Commission on Large Dams (ICLD). 1988. World Register of Dams. ICLD, Paris.

International Energy Agency (IEA). 1987. Clean Coal Technology. OECD Publications, Paris.

1991. E??ergy Statistics of OECD Countries. IEA, Paris.

Jensen, C., H. Price, and D. Kearnev. 1989. The SEGS power plants: 1989 performance. Pages 97-102 in A. H I anney and K. O. 1 und, eds. Solar Engineering 1989. American Society of Mechanical Engineering, New York.

Kellet, J. 1990. The environmental impacts of wind energy developments. Town Planning Review 61: 139-154.

Kennedy, T., J. Finnell, and D. Kumor. l 991. Considering environmental costs in energy planning: alternative approaches and implementation. Pages 145-150 in T. R. Mancini, K. Watanabe and D. E. Klett, eds. Solar Engineering 1991. American Society of Mechanical Engineering, New York.

Kohl, W. L. 1990. Methanol as an Alternative Fuel Choice: An Assessment The Johns Hopkins Foreign Policy Institute, Washington, DC.

Lipfert, F. W., R. G. Malone, M. L. Dawn, N. R. Mendell, and C. C. Young. 1988. Statisical Study of the Macroepidemiology of Air Pollution and Total Mortality. Report No. BNL 52122. US Dept. of Energy, Washington, DC.

Lynd, L. R., J. H. Cushman, R. J. Nichols, and C. E. Wyman. 1991. Fuel ethanol from cellulose biomass. Science 251: 1318-1323.

Maeda, Y., and M. Kai. 1991. Recent progress in pervaporation membranes for water/ethanol separation. Pages 391-435 in R. V. M. Huang, ed. Pervaporation 11 Membrane Separation Processes. Elsevier, Amsterdam, the Netherlands.

Matare, H. F. 1989. Energy: Fact and Future. CRC Press, Boca Raton, FL.

Mihlmester P. E., J. B. Thomasian, and M. R. Riches. 1980. Environmental and health safety issues. Pages 731-762 in W. C. Dickinson and P. N. Cheremisinoff. eds. Solar Energy Technology Handbook. Marcel Dekker,New York.

Moore, T. 1992. High hopes for high-power solar. EPRI Journal/ 17 (December): 16-25.

Ogden, J. M., and R. H. Williams. 1989. Solar Hydrogen: Moving Beyond Fossil Fuels. World Resources Institute, Washington, DC.

Peschka, W. 1987. The status of handling and storage techniques for liquid hydrogen in motor vehicles. International Journal of Hydrogen Energy 12: 753-764.

Pimentel, D. 1991. Ethanol fuels: Energy security, economics, and the environment. Journal of Agricultural and Environmental Ethics 4: 1-13.

Pimentel, D. 1992. Competition for land: development, food, and fuel. Pages 325-348 in M. A. Kuliasha, A. Zucker and K. J. Ballew, eds. Technologies for a Greenhouse-constrained Society. Lewis, Boca Raton, FL.

Pimentel, D., L. Levitan, J. Heinze, M. Loehr, W. Naegeli, J. Bakker, J. Eder, B. Modelski, and M. Morrow. 1984. Solar energy land and biota. SunWorld 8: 70-73 93-95.

Pimentel, D., M. Herdendorf, S. Eisenfeld, L. Olanden M. Carroquino, C. Corson, J. McDade Y. Chung, W. Cannon, J. Roberts, L. Bluman, and J. Gregg. 1994. Achieving a secure energy future: environmental and economic issues. Ecological Economics, 9: 201-219.

Schurr, S.M., J. Darmsradter, H. Perry, W. Ra msay, and M. Russel 1. 1979. Energy in America's Future: The Choices Before Us. A Study Prepared for RFF National Energy Strategies. The Johns Hopkins University Press, Baltimore, MD.

Sillman, S., and P. J. Samson. 1990. Impact of methanol-fueled vehicles on rural and urban ozone concentrations during a region-wide ozone episode in the mid-west. Pages 121-137 in W. L. Kohl, ed. Methanol as an Alternative Fuel Choice: An Assessment. The Johns Hopkins Foreign Policy Institute, Washington, DC.

Skinrood, A. C., and P. E. Skvarna. 1986. Three years of test and operation at Solar One. Pages 105-122 in M. Becker, ed. Solar Thermal Receiver Systems. Springer Verlag, Berlin.

Smil, V. 1984. On energy and land. Am. Sci. 72: 15-21.

Smith, D. R. 1987a. The wind farms of the Altamont Pass area. Annual/ Review' of Energy 12: 145-183.

Smith, D. R., and M. A. Ilyin. 1991. Wind and solar energy, costs and value. The American Society of Mechanical Engineers, IOth Annual Wind Energy Symposium: 29-32.

Smith, K. R. 1987b. Biofuels, Air Pollution, and Health: A Global Review. Plenum Press, New York.

Steele, A. 1991. An environmental impact assessment of the proposal to build a wind farm at Langdon common in the North Pennines, U.K. The Environmentalist 11: 195-212.

Tabor, H. Z., and B. Doran. 1990. The Beith Ha'arva SMWe solar pond power plant (SPPP)-Progress Report. Solar Energy 45: 247-253.

Tester, J. W., D. O. Wood, and N. A. Ferrari. 1991. Energy and the Environment in the 2 Ist Century. The MIT Press, Cambridge, MA.

Twidell, J. 1987. A Guide to Small Wind Energy Conversion Systems. Cambridge University Press, Cambridge, UK.

US Bureau of the Census (USBC). 1992a. Statistical Abstract of the United States 1992. US Government Printing Office, Washington, DC.

1992b. Current Population Reports. January ed. USBC, Washington, DC.

US Department of Agriculture (USDA).1992. Agricultural Statistics. US Government Printing Office, Washington, DC.

Vant-Hull, L. L. 1992. Solar thermal receivers: current status and future promise. American Solar Energy Society 7: 13-16.

Veziroglu, T. N., and F. Barbir. 1992. Hydrogen: the wonder fuel. International Journal of Hydrogen Energy 17: 391-404.

Voigt, C. 1984. Material and energy requirements of solar hydrogen plants. International Journal of Hydrogen Energy 9: 491-500.

Warner, I L. 1991. Consumer guide to energy saving windows. Solar Today 5: 10-14.

West, R. E., and F. Kreith. 1988. Economic Analysis of Solar Therma/ Energy Systems. The MIT Press, Cambridge, MA.

Williams, T. A., D. R. Brown, J. A. Dirks, K. K. Humphreys, and J. L. La Marche. 1990. Potential impacts of CO2 emission standards on the economics of central receiver power systems. Pages 1-6 in J. T. Beard and M. A. Ebadian, eds. Solar Engineering 1990. American Society of Mechanical Engineering, New York.

Winter, C.-l. 1986. Hydrogen energy-Expected engineering break-throughs. Pages 9-29 in T. N. Veziroglu, N. Getoff and P. Weinzierl, eds. Hydrogen Energy Progress Vl. Pergamon Press, New York.

Winter, C.-J. 1991. High temperature solar energy utilization after I 5 years R&D: kick-off for the third generation technologies. Solar Energy Materials 24: 26-39.

Winter, C.-l., and 1. Nitsch, eds. 1988. Hydrogen as an Energy Carrier: Technologies, Systems, and Economy. Springer Verlag, Berlin.

Winter, C.-l., W. Meineke, and A. Neumann. 1992. Solar thermal power plants: No need for energy raw material-only conversion technologies pose environmental questions. Pages 1981-1986 in M. E. Arden, S. M. A. Burley and M. Coleman, eds. 1991 Solar World Congress: Proceedings of the Biennial Congress of the International Solar Energy Society. Vol. 2, part 11. Pergamon Press, Oxford, UK.

Wittwer, V., W. Platzer, and M. Rommell. 1991. Transparent insulation: an innovative European technology holds promise as one route to more efficient solar buildings and collectors. Solar Today 5: 20-22.

Wolfson, R. 1991. Nuclear Choices. MIT Press, Cambridge, MA.

Worldwatch Institute. 1989. State of the World. Worldwatch Institute, Washington, DC.

1992. State of the World. Worldwatch Institute, Washington, DC.