• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Energy Conversion Goes Local: Implications for Planners

Posted on: Friday, 29 August 2008, 03:00 CDT

By Andrews, Clinton J

Problem: Emerging energy technologies ate bringing planners a new set of issues. The supply-oriented framework from engineering economics within which energy planning has ttaditionally been conducted may be useful for siting large refineries, power plants, and transmission corridors, but it is not helpful for mitigating conflicts at the site level, encouraging new technology adoptions, managing the demand for energy, or, especially, coordinating the diverse users of smaller, local energy facilities. Purpose: I provide an alternative conceptual framework for thinking about emerging energy planning tasks. I highlight factors not considered in the traditional model, and introduce terminology for characterizing key characteristics of the changing energy economy.

Methods: I draw on concepts from industrial ecology, urban metabolism, and ecological economics, and apply my new framework to a set of examples illustrating its advantages relative to the traditional approach to energy planning.

Results and conclusions: I propose that planners use network models to think about energy systems and focus especially on nodes where energy is converted from one form to another. Understanding the scale, scope, commodification, and agency of such nodes, and whether and when these attributes are open to change, can improve energy planning decisions for traditional energy investments such as power plants and for energy initiatives such as wind farms, rooftop solar systems, energy-efficient buildings, cogeneration, compact growth, and plug-in hybtid electric vehicles.

Takeaway for practice: Planners should do more than just mitigate energy facility siting conflicts. They should also identify points of governmental levetage on private decision makers, keep track of evolving technologies, bundle energy users with different temporal demand profiles, and help build smarter energy networks. Focusing on energy networks and their nodes should help planners see how they can be most effective.

Keywords: energy networks, infrastructure planning, energy efficiency, siting conflicts

Research support: This work was supported by the National Science Foundation and the New Jersey Board of Public Utilities.

A generation ago, planners helped site a centralized energy supply infrastructure of power plants, electric transmission lines, refineries, and pipelines. Recent attention has shifted to decentralized supplies and the effects of transportation, land use, and buildings on energy demand. It is time for planners to pay attention to the new spatial structure of energy systems.

This article does not dwell on the obvious links between energy and sustainability planning. The importance of good public policy, planning, municipal operations, and public outreach for sound energy choices cannot be overstated, but they have also been thoroughly covered elsewhere (Berke & Conroy, 2000; Campbell, 1996). This article also does not rehash the classic themes of community voice, noncoincident costs and benefits, and technological determinism that are well established in the siting literature (Hagman & Misczynski, 1978; Nye, 1992; Richardson, 1976/1991; Tauxe, 1995).

Rather, I develop a forward-looking conceptual framework for thinking about energy that relies on network models of complex systems. I begin by characterizing a traditional approach to energy planning, and then introduce a new, contrasting framework, which has roots in ecological economics. I show how the new framework usefully illuminates planning questions at the nexus of global warming, energy security, and infrastructure provision, and illustrate the framework with a set of examples.

Traditional Energy Planning

The traditional approach to energy planning is rooted in engineering economics, and its basic objective is to satisfy demand cost effectively. This approach favors large, centralized, supply- side solutions. It starts with an assessment of whether the current energy production capacity exceeds the current demand for energy by an adequate margin. It then looks forward to determine what existing capacity will be retired. Next, anticipated energy demand is forecast, which provides a basis for calculating how much new capacity is needed by which date. Then the planner selects the most cost-effective choice among a limited set of technological alternatives. Financing for the new capacity must be secured and, traditionally, energy utilities have been granted territorial monopoly franchises that provide stable revenue streams to secure their debt. Energy prices are set to cover depreciation, production costs, and a standard markup. Unlike regulated electric, natural gas, and district heating utilities described in the previous sentences, oil refineries rely more heavily on corporate equity financing and sell their products in minimally regulated markets.

The planner plays a major role in siting energy production facilities to minimize conflict and maximize economic efficiency. Building these facilities is a significant technological and logistical challenge, and regulating their operations to minimize safety and environmental problems is an ongoing task for planners and policymakers. This process is familiar to most infrastructure planners, and it worked well in an era of stable energy prices, growing demand, and plentiful sites. Stable energy prices made technology choices easy, and growing demand reduced the financial risks of overbuilding new capacity. Energy producers could aggressively pursue scale economies (reductions in long run average cost per unit achieved through expanding energy production facilities), given growing demand and, except for oil, a cost-of- service regulatory regime (Hyman, 1992).

However, the traditional energy-planning model began to fall apart during the 1970s as energy prices became volatile, demand growth sputtered, interest rates soated, and regulatory compacts disintegrated. Large, capital-intensive projects suddenly seemed risky (Caramanis, Schweppe, & Tabors, 1982). Siting challenges also increased, as host communities found their political voices, and as they began to feat stigma more than they valued compensation. The times changed and weakened the value of the traditional planning model. A set of emerging energy planning challenges has confirmed the need for a new approach.

Emerging Energy Planning Challenges

Since the 1970s, energy prices have continued to fluctuate dramatically under the influence of geopolitics, weather, and macroeconomic factors. Current high prices are associated with finite, geographically concentrated petroleum supplies, a growing appetite for commercial energy in Asia, a weak U.S. currency, and war in the Middle East. There is an underlying concern that energy supplies are no longer secure and adequate. Energy affordability is a corollary challenge. Electricity, natural gas, and petroleum products have become necessities of life in modern society. As prices rise, poor citizens and those on fixed incomes suffer. The third big challenge is global warming. Other environmental issues have affected the way energy industries operate but have not threatened their viability. Limits on greenhouse gas emissions will more fundamentally change the oil, gas, coal, and electricity industries.

I thus suggest considering alternatives to the traditional engineering model of energy planning. Rather than merely satisfying the demand for energy, I advocate using both demand management and supply procurement tools to equilibrate demand and supply. Likewise, rather than merely pursuing scale economies, I encourage planners to consider how to obtain economies of mass production (reductions in cost per unit obtained by producing many units) and economies of scope (per unit cost reductions achieved by co-producing multiple products) at many small, decentralized facilities, as well as economies of integration (per unit cost reductions achieved by changing the supply chain) across previously distinct sectors. I recommend we avail ourselves of technological opportunities to manage the risks associated with energy investments in a world with volatile energy prices, fluctuating demand, and impending carbon emissions constraints. Planners' traditional concerns are still salient as well, since all of these proposed facilities also need sites.

Promotion of efficiency, equity, and stability are the standard economic rationales for governmental intervention in the marketplace (Musgrave, 1959), and they are all valid concerns in energy markets. Hopkins (2001) defines four circumstances in which public plans can improve outcomes compared to a free market without public planning. Interdependence, wherein the result of one action depends on another and vice versa, lies at the core of facility siting and demand management problems. Indivisibility, wherein the size of the increment of action affects its value, characterizes the energy sector's pursuit of scale economies in large facilities. Irreversibility, wherein returning to a previous state incurs a very large cost, exactly describes the financial and environmental implications of building large power plants and refineries. Imperfect foresight, wherein actors acknowledge that more than one future is possible, describes the post- 1973 energy marketplace. Table 1 lists benefits planning could provide in circumstances displaying each of these attributes for seven types of energy initiatives I will use as examples throughout this article. Any improved approach to energy planning should accommodate today's challenges, employ defensible rationales, and provide useful insights on the relative efficacy of the emerging responses. A network model that focuses on energy conversion nodes achieves these ends.

A Network Approach to Energy Planning

Network models are familiar to planners studying global economic restructuring and cities. Infrastructure planners, environmental planners, and community development planners also rely on such models. Network representations of complex systems can make analysis tractable and highlight previously obscure features. In the energy economy, network branches typically represent energy or material flows. A common application portrays the supply-demand balance of the energy economy in terms of resource flows in (petroleum, natural gas, coal, renewables, nuclear) and user flows out (residential, commercial and industrial, transportation), all measured in a common metric as shown in Figure 1 . This whole-system perspective is familiar to environmental planners conversant in industrial ecology, urban metabolism, and fields allied with ecological economics (Andrews, 1999; Ayres, Ayres, & Warr, 2003; Kennedy, Cuddihy, & Engel-Yan, 2007; Newman & Kenworthy, 1999). A vector of technical coefficients that specifies how inputs are transformed into outputs in an input-output model of economic activity is another example familiar to planners, and similarly is most accurate if it reflects the unique aspects of the local situation.

Figure 1 does not show the flows that have no monetary value, such as the wasted energy from the many conversion processes that take place between the extraction of the primary resource and the moment of end use. A more complete model will capture the missing elements by focusing also on the nodes in the network, that is, the points at which flows transform into other flows, as shown in Figure 2. Castells' (1989) "spaces of flows" are primarily such sites of conversion.

It is the flows of energy and materials that have traditionally captured the most public attention. We worry about disruptions in the petroleum flows from the Middle East to the United States, electricity outages that darken our cities, and about the global warming caused by carbon dioxide flows from smokestacks and tailpipes into the atmosphere. Yet there are also conversion processes that turn, for instance, petroleum into gasoline, coal into electricity, and natural gas into carbon dioxide. Unlike the flows, conversion processes are often highly localized. They are clearly defined nodes in a complex network of energy and material flows. It is to these nodes that I want particularly to draw planners' attention.

A focus on conversion nodes will highlight energy facility siting issues, already a familiar concern for planners. A node is a transition point from energy input to output. The traditional energy facilities site map of New Jersey in Figure 3, for example, shows power plants, ports, refineries, transmission, and storage facilities. Yet as the technologies of distributed (rather than centralized) energy production and conversion become cost competitive and ubiquitous, new planning concerns arise. Site-level issues such as solar access are becoming important. Highly efficient cogeneration of heating, cooling, and electricity will only be economical in mixed-use developments. District energy systems serving clusters of buildings will depend on tight coordination of infrastructure and land use planning. New energy carriers and storage capabilities like hydrogen and advanced automobile batteries, respectively, will vastly increase the synergistic potential of the existing electricity, petroleum, and natural gas networks by allowing local conversions from one type of energy to another. Widespread adoption of more efficient end-use technologies (such as light bulbs) depends on effective public education and outreach.

Focusing on nodes is in fact a very subtle challenge. It forces us to consider questions of scale, scope, commodification, and agency, among others. These dimensions, discussed in greater detail below, frame the key issues and offer an appropriate simplification of reality for the particular questions and communicative contexts of urban planning. Table 2 shows how nodes differ on these dimensions across the same energy initiatives shown in Table 1 and described in greater detail later in this article.

Scale

What can be considered an energy conversion node? Every home light bulb is a node, converting electricity to light energy, but so is each household, workplace, town, state, and nation. The conversion node concept is useful at many different scales. Nodes in a global network may represent countries or trading blocs, whereas nodes in a municipal network might be schools, hospitals, or households. Nodes in a commercial building may be boilers and banks of lights, and nodes in an automobile may be the individual drive train elements. Top-down modelers, typically economists, tend to aggregate similar classes of objects into single nodes that probabilistically represent, say, households or industry, to which they attach strong behavioral assumptions. Bottom-up modelers, typically engineers, tend to disaggregate objects to a greater extent, creating deterministic representations of physical structures to which they attach strong but probabilistic assumptions about physical performance. For many energy economic questions it has become customary to create top-down/ bottom-up hybrid models that draw on the relative strengths of each approach. Network models of nodes and branches can then provide a unifying framework that bridges scales and analytical perspectives. Clearly there are nodes within nodes, implying linkages across scales, and possibilities for substituting detailed, highly localized models for probabilistic, highly aggregated nodes. Planners may need to find a middle ground perspective appropriate for the scale at which they work.

Commodification

The flows of coal, natural gas, and other energy sources are commodified in Figures 1 and 2; that is, they are represented using a common unit of measurement1 even though they have very different physical characteristics.2 This reflects the engineer's view that all of these are energy flows that can be reported in common units.3 These products are similar enough to substitute somewhat for one another in the marketplace, though barriers prevent seamless substitution. Which flows should be viewed as standard economic commodities is actually a strategic question, since substitutability depends on the network and nodes. Indeed, new conversion nodes could themselves become economically viable, interchangeable commodities (Geidl et al., 2007). Solar panels, gas turbines, light bulbs, and air conditioners are examples of energy conversion nodes that have already become standardized and are available only in specific sizes and types. Figure 4 provides an example of a new kind of node, a building that can also be called an energy hub because it expands opportunities for economies of scope by locating multiple energy conversions in one place.

Scope

Flows and transformations of a single standard commodity such as energy content can be depicted using simple networks (Figure 1), but it is also possible to track a matrix of interrelated flows such as energy and pollution, for example. With multiple flows, the transformation possibilities at each node become more realistic and yet more challenging to model because of their complexity. Figure 5 illustrates how such modeling could draw attention to an opportunity to reduce the long run average cost of each unit produced by expanding the scope of production to include one or more co- products (in this case, heat and electricity).

Agency

The rules governing how independent decision making is at a node, whether it is solitary or collective, and if the latter, whether adjudicated or negotiated, shape network adaptability, structure and efficiency, and thus are central to energy planning. Nodes may be passive points of conversion or decision points. Those that are decision points may represent decisions by individuals, such as a person choosing to turn a light on or off at home, or the aggregate decisions of multitudes, for example, the average fuel economy of the national automobile fleet. The question of agency is central for planners intending to influence outcomes, both for identifying who needs to be involved in each decision and what the key points of leverage on decisions will be.

Nodes as the Focus of Analysis

If planners are attempting to design better energy systems, considering these dimensions will lead to new ideas. Nodes can respond to price signals and other information. Nodes can be conversion sites for co-products such as heat and electricity. A node can be as small as a fuel cell in a basement or as large as a wind farm. New energy carriers like hydrogen or biodiesel bring new types of nodes, changing the relative attractiveness of centralized versus decentralized energy production. Energy conversion nodes are key elements in our energy future and they need planning.

Planning for Energy Conversion Nodes

Energy planning should include identifying governmental points of leverage on nodal decision-making agents, taking into account the likely future trajectory of technologies under consideration, addressing special siting concerns, and considering both how loads ate distributed over time, and how information flows affect demand and efficiency. Table 3 briefly summarizes how these concerns affect the same energy initiatives shown in Tables 1 and 2 and used as examples throughout. Public leverage consists of plans, regulatory approval processes, ordinances, financial incentives, and information ranging from product labels to use of the bully pulpit. The following sections discuss technological change, siting, load diversity, and information flows. Technological Change

Planners should consider how technological change affects long- lived investments in buildings and infrastructure because innovation is again taking place in the energy sector, after decades of stagnation. It is useful to first place the curtent eta in historical context.

History of Electrical Power. The U.S. electric power industry started in 1882 with small generators serving local centers. By 1910, the thermal efficiency of generating units (the percent of the fuel's energy converted into electricity rather than wasted as heat) had doubled to 8% and the average price of electricity had dropped to $1.71 per kilo- watt hour,4 one third of its price in 1890 (kWh, $2007). Between 1910 and I960 average prices fell below $0.17 per kWh ($2007) primarily as a result of scale economies, as the maximum capacity of power generators jumped from 10 MW to 1,000 MW, and thermal efficiencies climbed to near 40% in the best units (Hirsh, 1989, pp. 4, 5, 9). Since 1970, however, conventional generating units have not gotten larger (1,300 MW) , and thermal efficien- cies have also been unchanged (40% in the best units and 33% for the average plant), holding prices constant in real terms (Hirsh, 1989, pp. 4, 5, 9). Coal and oil were both used extensively for electricity generation until 1973, when price shocks due to the Arab oil embargo and the Iranian revolution made oil expensive, and since that time most electricity has been generated by burning coal. The first commercial nuclear-powered electrical generating plant came online in 1957, and the most recent in 1996, with operating efficiencies remaining relatively unchanged over that period even as their cost-effective scale grew from 68 MW to 1121 MW (American Society of Mechanical Engineers, 1980; Disosway 2006; Energy Information Administration [EIA], 2005). Such large nuclear plants proved difficult to finance and integrate into the grid.

As noted above, the net scale economies in conventional electricity generation were exhausted by the late 1970s, and the average size of units has trended downward since then (Dunsky, 2000). Only during the 1990s did the widespread adoption of gas- turbine combined-cycle technology (which begins by burning natural gas or oil in a combustion turbine and then drives a steam turbine with the leftover heat) raise thermal efficiencies to 50% (EIA, 2000, 2007a). Gas-turbine combined-cycle power plants are smaller than conventional coal-fired plants (typically in the 250 MW range) and use mass produced components. These two characteristics reduce investment risk and allow building in phases, increasing the electricity supply in small increments in response to demand. However, these plants require natural gas or light oil as fuel, and skyrocketing fuel prices since 2003 have made them unattractive compared to coal-fired power plants (EIA, 2007b). Nonetheless, current real electricity prices are low by historical standards, with the 2007 U.S. average electricity rate at $0.09 per kWh (EIA, 2007c).

Alternative Technologies: Solar and Wind. Public policy stimulated markets for solar photovoltaic electricity (using arrays of mass-produced cells or thin films to convert light from the sun directly into electricity), and worldwide panel manufacturing rates increased from a few kW in 1976 to 2,400 MW in 2006, while prices for solargenerated electricity dropped from $5.00 per kWh to $0.50 per kWh ($2007; Solar Energy Industries Association, 2007; Solar Energy Technology Program, 2007). Clearly these prices will have to fall further to make solar electricity widely competitive.

Wind power has done better commercially. Average wind turbine sizes increased from 50 kW in 1982 to 1,600 kW in 2006, while overall manufacturing rates jumped from under 100 MW to 15,000 MW annually worldwide and prices dropped from $0.20 per kWh to $0.04 per kWh ($2007) during this period, making wind power competitive5 with conventional electricity generation (Cory, Bernow, Dougherty, Kartha, & Williams, 1999; Wiser & Bolinger, 2007).

Several states have adopted aggressive demand-stimulus policies for renewables. In New Jersey, for example, utilities are obligated to acquire 22.5% of electricity from renewable sources by 2020 (New Jersey Board of Public Utilities, 2006).

Cogeneration. One of the most interesting alternative technology trajectories is that of cogeneration of combined heat and power. Finding a use for what otherwise would be waste heat raises total conversion efficiency to 75% or more (Casten, 1998). First- generation cogeneration plants burned coal or oil to produce high- pressure steam for process heating and also for generating electricity in a steam turbine. This technology yielded scale economies, but found only modest use, mostly in large downtown utility districts and mammoth industrial plants. In the United States, monopoly franchise regulation of utilities limited the spread of the technology, so that very little was invested in such systems from 1930 until 1978 because their returns were not as attractive as electricity-only investments during the era of rapid suburbanization (McCaughey, 1998; Pierce, 2004).

Federal legislation changed the regulatory rules in 1978, when the Public Utilities Regulatory Policies Act opened the door for nonutility, third-party generation of electricity.6 This policy spawned a variety of technological innovations in renewable energy and small-scale power generation. A new breed of cogenerators appeared, this time generating electricity from large multi- megawatt internal combustion engines, and then capturing the waste heat for process use (EIA, 2000). Beginning in the 1990s, gas turbines started replacing big diesel engines in these applications, and cogeneration became widespread for university campuses, governmental sites, and hospital complexes with electricity demand of 10 MW or more. Very recently gas microturbines (on the order of 50 kW) have made cogeneration economically feasible within single buildings (California Energy Commission, 2007). Future progress should yield affordable fuel cells or turbines at the scale of a single household (5 kW).

Oil and Gas. Investment in oil and gas refineries, distribution, and retail facilities has been very modest for the past 50 years. For example, New Jersey's six refineries cluster on industrial sites used for such purposes since the 1880s (EIA, 2007d). Siting difficulties, environmental liabilities, and volatile profit margins have reduced the attractiveness of refineries as investments. Instead, favorable tax policies have encouraged innovation in new exploration and extraction technologies that have allowed oil producers to push the depletion horizon for these valuable fossil fuels further and further into the future.

Consumer Technology. Innovation has also affected consumer demand for energy. Compact fluorescent light bulbs entering the marketplace since 2000 are four times more efficient at delivering lumens per watt than incandescent bulbs (EnergyStar, 2008). Boilers and water heaters using condensing pulse combustion technology operate at 95% combustion efficiency compared to 65-85% for conventional technologies (Council of Industrial Boiler Owners, 2003; EnergyStar, 2007). A variety of commercially available automotive technologies are now achieving over 50 miles per gallon compared to the U.S. fleetwide average of 20 miles per gallon (Federal Highway Administration, 2005).

Thus, technological innovation comes in spurts, can be driven by public policies, and makes new configurations of the built environment feasible. Economies of scale are not the only possibility, and substantial progress in reducing energy prices and increasing opportunities are due to economies of mass production, scope, and integration. Small, mass-produced alternatives now include gas turbines, wind turbines, solar panels, and low-energy light bulbs. Cogeneration offers scope economies. Plug-in hybrid vehicles offer integration economies. Large-scale, public utility enterprises are needed less to produce the energy than to deliver it. The relevant energy conversion nodes are becoming smaller, more ubiquitous, more integrated into the landscape, and more subject to local influence.

Siting Challenges

Power plants and refineries are classic industrial land uses that planners routinely separate from residential areas to minimize the impacts of noise, dust, fumes, and heavy transport. These energy conversion nodes are unpleasant but there are not very many of them. Whether emerging technologies belong in the same category depends on the spatial spillovers each technology produces.

Wind turbines emit no pollution but they loom large on the landscape, with current technology (1 MW turbines) taller than a football field is long. Like electric transmission towers, wind farms are visible from miles away and therefore provoke strong feelings from residents concerned about the aesthetics of their viewshed. Wind farms are compatible with many agricultural and some industrial land uses where neighbors value the economic benefits and do not mind the visual impacts. Residents resist wind turbines in tourist areas such as Cape Cod and New York State's Adirondack Park while welcoming them in adjacent agricultural counties.

Rooftop solar panels for generating electricity or hot water depend on direct access to sunlight. Shading by nearby structures is a significant threat that good planning can mitigate. Standard tools to guarantee solar access include subdivision regulations and zoning codes governing building height and setback requirements. Large- scale cogeneration systems, like conventional power plants, belong in industrial or institutional zones because their noise and smokestacks make them unsuitable for residential neighborhoods. However, small-scale cogeneration using a microturbine, for example, takes place entirely within a building and imposes no more impacts on neighbors than a conventional boiler or furnace. The regulatory issue for these units is operational safety rather than spatial spillovers, meaning this technology should be scrutinized by fire and building code officials rather than planners.

District energy systems distribute energy from a cogeneration plant to nearby buildings using piped underground loops carrying hot water, chilled water, and electric power. Such systems offer substantial energy efficiency gains and save on construction costs in buildings by allowing owners to forego boiler rooms. Planners can influence the adoption of this technology because it requires access to public rights of way for piping systems, coordination among investors and multiple building owners to create this shared utility, and selection of users with adequate load diversity to make the system economically feasible.

Temporal Profiles

Infrastructures of all types face the challenge of meeting time- varying demand. Roads and rails are congested during rush hour and underutilized during off-peak hours. Sewer systems built for regular use are overwhelmed by stormwater loads; hence modern systems separate sewage and stormwater. Planners devote much effort to estimating the demand for parking and have developed a detailed understanding of the temporal demand profiles for residential, commercial, and other land uses, pursuing complementarities to ensure higher utilization of this expensive investment. The economics of capital-intensive energy systems are also highly sensitive to capacity utilization and thus the temporal profiles of demand and supply.

Pursuit of greater load diversity7 was one of the driving forces behind the widespread interconnection of electric power networks in the early 20th century and most reductions in average price during this period were achieved through economies of interconnection rather than economies of scale (Hirsh 1989, pp. 4, 9). Fewer power plants could more economically serve a larger number of customers once individual loads with diverse schedules were aggregated.

Cogeneration and district energy systems also apply this principle, using capacity efficiently by aggregating diverse loads. By combining residential loads (mostly evenings and weekends) and office loads (mostly weekdays), for example, cogeneration becomes much more costeffective. New technologies such as plug-in hybrid electric vehicles provide further opportunities for load leveling. Figure 6 shows the temporal profiles of different energy users and illustrates how complementary user types improve load diversity. Packaging local uses to create a viable energy conversion node is an important new role for planners.

Solar and wind technologies deliver energy intermittently, and their temporal profiles of supply strongly affect their economic attractiveness. Because their energy production cannot be centrally controlled, or dispatched, utilities often discount their value. Absent time-of-day rates, the valuable coincidence of the solar supply and system-wide demand peaks goes unrewarded.

Information Flows

Information about energy-network relationships is increasing. Traditionally, electricity and natural gas customers were charged regulated prices, billed monthly, which gave them almost no information about the true cost of their energy use, and was made even less relevant by being greatly delayed. Today, large customers see time-varying prices one day ahead or even in real time, and can adjust their usage accordingly. Real-time pricing is beginning to penetrate the residential market too. This increased information allows customers to respond to price signals and invest in technical solutions that save them money. It also makes management of the grid much easier.

In many states, utilities are also now required to tell customers about the environmental impacts of energy, using bill inserts that report carbon dioxide, sulfur dioxide, and other pollutant emissions per kWh consumed. Where regulators allow retail choice, this information provides a basis for differentiating greener energy products from conventional products. In these and other ways, the information asymmetries between energy producers and consumers are diminishing. But informing decision makers at energy network nodes at every scale remains a key planning and policy task.

Examples

The examples contained in Tables 1-3 are discussed below at greater length, to illustrate why the factors in the foregoing discussion are relevant and how they interact with one another. The traditional centralized energy system serves as the point of departure. Planning questions regarding wind farms, rooftop solar panels, energy efficient buildings, cogeneration, compact growth, and plug-in hybrid vehicles illustrate what planners can learn by focusing on energy conversion nodes.

Traditional Energy System

The elements of the traditional energy system important to traditional energy planning are the few big nodes (power plants and oil refineries) and the flows that use electric transmission lines and oil or gas pipelines. Of these, the transmission lines have been most problematic because they cut across huge swaths of landscape, impact many property owners, and provide minimal benefit to the people who bear the aesthetic costs. For example, the existing electricity grid in New Jersey is part of a large regional market called the Pennsylvania-Jersey-Maryland Interconnection (PJM) that extends west to Illinois and south to North Carolina, and contains 1,271 power plants and 56,250 miles of high voltage transmission lines owned by 17 utility companies serving 51 million people (PJM, 2007a). PJM has just launched a highly controversial process to site a major new transmission line through or near the scenic Delaware Water Gap National Recreation Area (PJM, 2007b). Siting power plants and refineries is still a planning challenge, but impacts are more localized, compensation is much more feasible, and Euclidean zoning is more adequate to mitigate conflicts than is the case for siting transmission lines.

As summarized in Table 2, the scale of nodes in a traditional electricity network is very large, but has been shrinking, from 1000 MW units to much smaller power plants. Their scope is limited to a single energy type, electricity. The electrical energy itself is the commodified part of the production chain. Key actors include the corporate owners of the power plants and transmission and distribution networks.

Table 3 shows the key planning issues for a traditional energy system. The state public utility regulator is the strongest point of leverage on the owners of the power system, although environmental regulators and federal public utility regulators also have influence. Siting issues that affect the traditional energy system fit well within a classic doctrine of separating industrial and residential uses. The core technologies are mature, even stagnant. Load diversity has historically been pursued by extending the grid to include more and more customers, a strategy that works well but exposes the network to widespread, cascading outage events sometimes called "normal accidents" (Perrow, 1984). Information flows are slow and asymmetrical, with consumers learning their costs only in a monthly utility bill. This gap imposes personal and social costs because consumers cannot adapt to changing conditions, cleaner technologies are devalued, and utility companies must overbuild to meet the nonresponsive demand.

Table 1 shows how Hopkins' (2001) rationales for planning relate to the traditional power system. Planners should coordinate energy facility siting because of its interdependence with other land use choices. Regulators should create monopoly franchises or governmental ownership to finance indivisible investments in large power plants and transmission systems. Planners should scrutinize proposed, long-lived power plant investments because of their irreversibility. Planners should weigh carefully how to balance future energy supply and demand because they have imperfect foresight.

Approaching energy planning by focusing on energy conversion nodes highlights how the traditional energy system fails to pursue economies of scope, how little it has been advancing technologically, how it hampers decision making by providing only minimal information, and how it overlooks the implications of uncertainty about the future. The scale of these utility networks is so vast that municipal planners rarely have to worry about them, but that is not the case for the remaining examples.

Wind Farms

Pursuing cleaner, more secure energy supplies, governments worldwide have encouraged wind-power generation. In the United States this has led to a growing number of wind farms such as the one in Lowville, New York (shown in Figure 7). The site currently has 120 turbines, with 75 more turbines planned, and they are spread out across 21,000 acres of farmland on sites that total about 200 acres (Maple Ridge Wind Farm, 2006). Each turbine provides lease payments to local landowners of $5,000 to $10,000 per year and delivers 1.65 MW of electricity whenever the wind blows at appropriate speeds, enough to supply the electricity needs of 500 homes when accumulated on an annual basis (DePalma, 2006). Wind farms have many similarities to traditional power plants but also a few key differences, highlighted in Table 2.

The scale of nodes is increasing for wind farms, increasing both the size of individual turbines and of wind farm sites. The scope of a wind farm is limited to a single energy product, electricity, but a wind farm site can support additional agricultural activities including animal grazing and growing crops. The wind farm delivers commodity electricity, but the wind turbines are also a mass produced item. Agency resides with the wind farm developer, but local planning boards and state utility and environmental regulators can strongly influence decisions. Table 3 shows that the key planning issue is the siting of the wind farm, which has four distinct aspects: Is the wind resource adequate? Is there nearby access to the electric transmission grid? Is the site on a bird migration route or in a bat feeding area? Do the neighbors object? Addressing the concerns of neighbors often requires lengthy public processes and the use of sophisticated view-shed visualization techniques. Wind technology is mature but still improving fairly rapidly. Because the wind blows intermittently, the temporal profile of wind generation reduces its economic attractiveness to power purchasers. Wind farm and utility grid operators are actively seeking out better daily and hourly local weather forecasting information to anticipate and accommodate the variation in a wind farm's electricity output.

Rationales for planning wind farms, summarized in Table 1 , include minimizing siting conflicts, achieving a large enough size to optimize operational expenses and justify transmission line investments, choosing designs that take into account the potential for technological obsolescence, and managing the uncertainty over energy prices and compatibility with neighboring land uses in the future. By focusing on the wind farm as an energy conversion node, planners can identify the relevant scale issues, the compatibility with agricultural land uses, the implications of betting on an evolving technology, the economies associated with mass production of turbines, the challenge of capacity to generate electricity being only intermittent, the special weather information needs, and the key decision makers affecting the adoption of this clean -energy solution.

Rooftop Solar Panels

With solar, instead of planning to minimize the conflicts caused by a few big nodes, planners face a different problem. Figure 8 shows a house in Highland Park, New Jersey, that has a solar photovoltaic array installed on its roof. Rooftop solar is a distributed energy generation system that collects sunlight and transforms it into electricity or hot water on thousands of discrete rooftops owned by individuals. (There are also a few centralized solar power plants located in desert areas.) Building owners are unlikely to invest in rooftop solar if there is a significant risk that the panels may become shaded in the future, a matter than often depends on the actions of an adjacent landowner. Numerous municipalities have enacted solar access ordinances that (a) encourage new subdivisions be designed to avoid solar conflicts, and (b) protect the solar access of existing rooftop solar systems on an as-requested basis once they are built (National Center for Appropriate Technology, 2004; O'Neill, 1986). In the Highland Park case, no such law yet exists, leaving the solar array vulnerable.

In the case of rooftop solar systems (see Table 2), the scale of the node is the individual building, its scope remains a single energy type (electricity), the commodified part of the production chain is the solar installation, and the building owner is the key actor. Local zoning laws, state electricity tariffs, and federal tax credits all influence the building owner. The key planning issues (see Table 3) include preservation of solar access, the emerging status of this immature technology, the incentive to rely on the grid for backup power rather than expensively serving every hour of demand, day or night, and the dramatic economic benefits owners can obtain by substituting solar power for their most expensive electricity when smarter electric meters provide them sufficient information. Strong rationales for planning (see Table 1) include minimizing conflicts among adjacent property owners, serving public objectives to reduce pollution and enhance energy security, encouraging property owners to adopt longer-term investment horizons, and preserving the option to use solar in the future as its cost drops.

Rooftop solar systems are invisible to traditional energy planners, showing up merely as a reduction in system-wide demand. A focus on energy conversion nodes allows planners to design strategies to support the successful adoption of this technology.

Energy-Efficient Buildings

Some energy conversion nodes are so decentralized that regulations, price signals, and public education are the most effective ways to improve them. These are the nodes of functional end use shown on the right side of Figure 2. California's Energy Code (Part 6 of the Title 24 Building Standards Code) contains energy conservation standards applicable to all residential and nonresidential buildings in the state. It dates back to the 1970s, and was most recently updated in 2005 (Division of the State Architect, California Department of General Services, 2007). It transformed architectural and engineering practice within a short period of time and made California's buildings among the most energy efficient in the nation. Along with high energy prices and a mild climate, the Energy Code has helped California drop to 48th place among the 50 U.S. states in per capita energy consumption, and although it has 12.2% of the U.S. population, it consumes only 7.3% and 8.8% of the nation's residential and commercial energy, respectively (EIA, 2007e).

Other locations are pursuing voluntary approaches. For example, the New Jersey Meadowlands Commission is a regional planning entity for a large area in northern New Jersey. It is building a new education center and has chosen to make it a green building certified under the Leadership in Energy and Environmental Design (LEED) rating system of the U.S. Green Building Council (2007).

The domestic construction industry is very conservative, and it remains difficult to find architects, engineers, contractors, and financiers who are willing to experiment with new approaches to the design of long-lived buildings (Koebel, 2007). Yet the global building industry has grown inventive, driven by an Asian high-rise building boom and a European environmental ethic, creating a growing gap between standard practices in the United States and global best practices.8 This is particularly true in the areas of energy efficiency, water efficiency, and indoor environmental quality.

Governmental regulation tends to drag laggards forward rather than encouraging the leading innovators. For example, building codes and appliance efficiency standards set minimum performance levels. These standards are ratcheted up only rarely. At the national level there were reforms in the 1970s and early 1990s, while at the state and local levels, especially in rural areas, many building codes have not changed much since the 1960s. By contrast, voluntary rating systems such as EnergyStar and LEED have helped codify best practices and make them understandable to building users. Planners can aid both the regulatory and voluntary efforts to improve building energy performance by updating codes, encouraging adoption of green-building ordinances, and writing aggressive performance standards into redevelopment plans. The Meadowlands Commission is hoping to use its LEED building as an exemplar that will reduce local resistance to green buildings, while building a statewide coalition to change building codes.

In energy-efficient buildings (see Table 2), the scale of the node is the individual building and its equipment; its scope is potentially broad, and may include coproducing energy for end uses such as lighting, heating and cooling; the key commodities have been the energy-using equipment, although processes such as EnergyStar and LEED certification are attempting to standardize building practices too; the key actor is the building owner; and those with the strongest influence on the owner are local building inspectors, contractors, and equipment suppliers. Planning issues (see Table 3) include identifying accessible locations, optimizing building orientation and landscaping; choosing emerging technologies that are also cost effective; correcting the frequent oversizing of heating and cooling equipment to accommodate the peak day's loads; and seizing the emerging opportunity for smart buildings to manage demand in response to energy price signals. Defensible rationales for planning (see Table 1) include the need for pump-priming in emerging markets for green technology, inadequate information for many parties to make good marketplace decisions, the tendency of building owners to adopt short-term investment horizons, and a need to set performance standards for buildings and equipment that reflect the likely range of future energy prices.

As with the solar example, a traditional energy-planning approach sees energy efficient buildings as nothing more than reduction in demand. Focusing on energy conversion nodes instead encourages planners to address site selection and design, take technology change into account, encourage energy storage and demand management with clearer price signals, couple smart buildings with a smart grid, and educate and inform building designers, owners, occupants, regulators, and equipment suppliers.

Cogeneration

Pursuing economies of scope at an energy conversion node requires tight integration of multiple systems and actors, often near the point of functional end use in Figure 2. For example, Oklahoma City, Oklahoma, has a districtenergy system operated by Trigen Corporation that provides centrally produced steam, hot water, and chilled water to approximately 20 commercial, governmental, institutional, and hospitality customers in the central business district and also cogenerates electricity. The distribution systems extend about 1 mile, and deliver up to 300,000 pounds of steam per hour, approximately 17,600 tons of chilling capacity,9 and electrical power production of 1.2 megawatts (Trigen, 2007). Cogeneration, or the combined production of heat and power (see Figure 5), is a technology little-known to most planners, yet it is widely used in college campuses and other large facilities. Cogeneration is not confined to the cold North because absorption-chiller technology10 makes it possible to turn steam and hot water into chilled water for use in cooling buildings, ensuring a year-round demand for the heat and electricity. Absorption chillers use heat instead of an electric compressor to drive refrigerant through the thermodynamic stages of the refrigeration cycle. This technology predates current electricity-driven cooling technology, but innovations and pursuit of scope economies have recently increased its economic attractiveness. Cogeneration is financially viable only when the temporal profiles of demand for heat and power ensure high capacity utilization. Since the laws changed in 1978, electricity can always be sold back to the grid, so the key challenge is to find a steady demand for the cogenerated heat. Location becomes important because heat cannot be transported over long distances. College campuses and military bases work well because they have complementary residential, classroom, laboratory, and administrative heating and cooling needs close to one another. Hospitals and prisons also work well because they heat and cool around the clock. A cogeneration plant is often the prime mover for a district-energy system. For example, the Oklahoma City system serves over five million square feet of commercial space, numerous federal and county government facilities, and one million square feet of hotels and other hospitality facilities. Implementing cogeneration on a college campus is a relatively straightforward engineering and financing challenge because it serves a single owner. Cogeneration as part of a district-energy system must also coordinate multiple building owners and the public officials who control access to under-street utility corridors, interdependence making planning essential. This technology is evolving toward the more universal concept of the energy hub mentioned earlier.

As shown in Table 2, cogeneration operates at scales ranging from the individual building to the campus or subdivision, its scope explicitly includes a range of energy types, the commodified elements are the prime movers such as gas turbines, the key agents are system owners, and those with the most influence on the owners include planning officials and state public-utility regulators. Major planning issues (see Table 3) include achieving an economically optimal mix of uses to ensure a viable temporal load profile, choosing an appropriate system size given that the technology is rapidly becoming feasible at smaller and smaller scales, deciding whether the economic advantage of serving multiple buildings is worth the coordination challenge, and developing operating strategies that reflect timevarying energy prices. Reasons for planning cogeneration systems (see Table 1) are that such systems require both an economically viable scale of operations and the load diversity likely to result from multiple users located in close proximity, and that such systems must manage future uncertainty with a long-lived, tightly coupled compact binding together multiple buildings and users and a plan that remains attractive as energy prices fluctuate.

Cogeneration was uncommon until federal laws changed in 1978 to encourage nonutility generators. Even since then, it has been limited mostly to college campuses, military bases, and hospitals where one owner controls all of the facilities. A focus on energy conversion nodes, especially near the points of end use, heightens our appreciation of the value of this technology while also identifying difficult coordination issues that planning must overcome before widespread adoption will occur.

Transit-Oriented Development

Settlement patterns influence energy demand. For example, transit- oriented development (TOD) is gaining popularity as a way to reduce gasoline consumption. Assuming vehicle ownership is a good proxy for household transportation-related energy consumption, the 2004 national average is 1.8 vehicles per household (U.S. Census, 2007), whereas an Evanston, Illinois, TOD averages 1.1, the average across a dozen California TODs is 1.4, and the average across 16 New Jersey transit villages is also 1.4 (Cervero et al., 2004; Nelson-Nygard, 2006; Wells & Robins, 2006). Yet it is surprisingly difficult to demonstrate the expected links between settlement patterns and energy use in the United States. People living in walkable neighborhoods with transit access do reduce vehicle miles and thus energy consumption compared to the national average, but outside our largest, densest cities this difference is modest, on the order of 10% (Kuzmyak, Pratt, Douglas, & Spielberg, 2003). People seem to shift to transit when housing density within a one-half mile radius of the rail station reaches about 3,000 units, with employment density, parking, and attractive design also playing roles (NelsonNygard, 2006). Newman and Kenworthy (2006) claim greater energy savings for TODs based on international data, but identify a similar density threshold of about 10,000 people and jobs widiin a 10-minute walking radius. Housing units in dense downtowns are often smaller than those in the suburbs, and more often share common walls, both of which reduce energy use. A countervailing force is the urban heat island effect that increases net energy use downtown because air conditioning loads increase even as heating loads decrease (Stone, 2005). Compact growth seems to deliver a modest net reduction in per capita energy use (Rong, 2006), but to date the effects are more incremental than transformative in the U.S. context.

As summarized in Table 2, transit-oriented development has a neighborhood scale, has a broad scope that includes the full range of energy end uses, commodifies both energy flows and the style of housing, features the developer as the primary actor, and identifies the planning board as the governmental body most able to influence the developer. Key planning issues (see Table 3) include achieving walkability and a good mix of uses, contributing to regional improvement, and sending clear signals to developers that compact growth is desired.

Rationales for planning (see Table 1) include the dual coordination challenges of attractively mixing residential with other uses and achieving concurrent investments in infrastructure and private development; serving the public good by reducing pollution and improving public health; and making farsighted design decisions so that settlement patterns remain attractive under a variety of future energy and transportation system scenarios.

Traditional energy planning treats neighborhoods as sources of energy demand to be satisfied. Many of the enduse energy conversion nodes highlighted in Figure 3 exist within neighborhoods, in houses, cars, streetlights, and other points that deliver functional services. Compact growth that gets people out of their cars particularly reduces energy use, as well as serving other longstanding planning objectives. However both mass transit and cogeneration succeed only above density and diversity thresholds that are high by current standards.

Plug-in Hybrid Electric Vehicles

With decentralized energy conversion nodes, it becomes possible to integrate previously separated energy systems. The California Cars Initiative (CalCars) and the Mid-Atlantic Grid Interactive Cars (MAGIC) partnership are efforts on opposite coasts to create the commercial climate and the information and communications infrastructure needed to support plug-in hybrid vehicles like the one shown in Figure 9 (CalCars, 2007; Mid-Atlantic Grid Interactive Cars, 2007). Hybrid vehicles that include both a gasoline-fueled internal-combustion engine and a battery-powered electric motor that doubles as an electric generator during regenerative braking are becoming quite familiar. Plug-in hybrids, by contrast, run on battery power for local trips and gasoline for longer trips. The hybrid is significant because it represents a technological bridge to several attractive longer-term automotive futures, including the clean, potentially non-fossil-fuel dependent, all-electric car and the hydrogen fuel-cell car.

The plug-in hybrid has more battery storage than a conventional hybrid, and it is equipped to recharge its batteries from the electric grid instead of relying exclusively on the internal- combustion engine and regenerative braking. This connection between the vehicle and the electric grid opens up possible synergies. Not only can the vehicle buy power from the grid to recharge its batteries, but it can also sell power back to the grid by discharging those batteries. Since many vehicles sit unused for most of the day and night, they represent an underutilized asset. The time many vehicles sit idle between the morning and evening commute coincides with the peak hours for demand on the electric grid, while the hours they are idle at night correspond to the time when electricity use is lowest. Plug-in hybrids can help level out the electricity-demand profile and earn money for doing it. In areas where increased amounts of wind and solar energy are coming online, this storage capacity will become particularly valuable to help keep the grid stable. Wind and solar energy generation fluctuates with the weather and time of day. Engineeringeconomic studies suggest that vehicle owners should see a net financial gain on the order of $1,000 per year from buying power off-peak and selling it back on- peak (Kempton & Tomic, 2005). None of this will occur unless public utility commissions and utilities allow such transfers, structure the sell-back rates fairly, and support the development of an adequate information infrastructure for carrying out these many decentralized transactions. The scale of the plug-in hybrid vehicle node (see Table 2) is the vehicle and its associated parking spaces, its scope includes the provision of both mobility and electricity, the commodified elements include the electricity delivered and the vehicle itself, the most important actor is the vehicle owner, and the governmental bodies that are most influential include the state public utility and environ- mental regulators. The planner's direct role in supporting the diffusion of plug-in hybrids (see Table 3) includes ensuring that many parking places at home and at work have electric outlets and wireless internet access for carrying out transactions, evaluating the progress of emerging battery technologies, and ensuring that time-of-day electricity price signals encourage vehicle users to charge and dis- charge their batteries at times that improve the overall electric system load profile. Traditional energy planning treats transportation system energy demand separately from the buildings and industries served by the electric power sector, but focusing on energy conversion nodes highlights vehicle-to-grid synergies and provides a basis for evaluating whether these innovations are yet attractive enough for widespread deployment.

Strong rationales for planning (see Table 1) include the need to coordinate the provision of suitable vehicles, grid access, and smart metering; pursuing the public goods of better grid stability, lower pollution, and improved energy security; making major, long- term investments in smart meters, new pricing regimes, and vehicles; and designing pricing and metering systems that work regardless of whether the automotive future belongs to plug-in hybrids, all- electrics, fuel-cell vehicles, biofuels, or some other option.

Discussion and Conclusions

Energy is important to Americans on many levels, yet has traditionally been linked to planning practice only by the siting of energy facilities, especially power plants, refineries, and electric transmission lines. However, the innovation spawned by utility deregulation, economic globalization, and spillovers from the high- tech sectors, and the sense of urgency sparked by fears of global warming, oil wars, and resource depletion have created new opportunities for planners to engage with energy questions.

Decentralized power-generation and storage capabilities are enjoying increased private investment and governmental support, and these technologies will increasingly complement the existing energy- supply networks. Planners will be called on to resolve siting conflicts to preserve solar access. They will also have a role in encouraging green-building practices. They will help make cogeneration and districtenergy systems feasible by packaging users with complementary energy-demand profiles. They will encourage compact settlement patterns that use energy more efficiently. They will accommodate dramatic technological innovations in the transportation sector.

Utopian writers see these changes to energy systems in grand terms, predicting that they will decentralize the economy, democratize decision making, empower individuals to be both consumers and producers, and break the faceless utility monopolies (Droege, 2006). Less optimistic observers argue that we will fail to seize the moment and instead slip into a long emergency of global warming, oil depletion, and suburban decay (Kunstler, 2005). Dramatic changes are indeed becoming possible, but they will progress far more quickly if planners get it right on the ground.

The soft-energy revolution (Lovins, 1976) planted seeds that are only today germinating in the form of energyefficient light bulbs and decentralized energy production. To the extent that "civilization advances by extending the number of important operations which we can perform without thinking about them" (Whitehead, 1911, p. 61), it is important to recognize how the grand connects to the prosaic in energy planning. Planning is the application of foresight to action. It is time to rethink some of our planning operations because of the emergence of smarter energy networks with more decentralized and imaginatively integrated energy conversion nodes. These nodes are small in scale, multidimensional in scope, are themselves commodified, and are subject to localized, individual decision making. They are right where planners work, in the nation's neighborhoods.

Planners should also focus on nodes because they are the points where we convert flows and therefore have great leverage on the overall energy economy. These conversions provide opportunities to improve system-wide energy efficiency and capture co-product synergies.

This move toward distributed energy is, of course, somewhat counterintuitive. One of the attractions of large, centralized energy nodes is that they are easy to manage, being few in number and typically owned by a single entity. Also, for society to be better off, economies of scope in cogeneration must exceed economies of scale in central power plants, under a range of future fuel- price trajectories. It is possible to constrain future choices too tightly, as happened when factories in the former USSR foundered during the 1990s and the residential buildings with which they shared inadaptable disttict heating systems lost their heating in the middle of the cold, Russian winter. So, rather than dogmatically embracing decentralized solutions, planners will better serve society by considering each case on its merits, and recognizing that these new, localized nodes generally connect to and complement the existing system.

Acknowledgments

Frank Popper, Amy Helling, and several anonymous reviewers made valuable suggestions that improved this article.

Journal of the American Planning Association, Vol. 74, No. 2, Spr

Source: American Planning Association. Journal of the American Planning Association

More News in this Category