Energy supply chain and operations risks and vulnerabilities
Climate change may affect the urban energy supply chain in three principal ways: through impacts on primary energy feedstock production or supply networks delivering these feedstocks to power plants; impacts on power generation operations; and via impacts on the energy transmission and distribution network. Our understanding of these risks varies widely, as does the severity each risk presents to cities around the globe.
Energy resource production and delivery
Primary energy fuel stocks tend to be found away from urban areas, but the impacts of climate change on the sourcing and processing of these materials would nonetheless be felt in urban areas, albeit in an indirect manner as cost impacts ripple across national or global economies.
For example, the US Climate Change Science Program (Bull et al., 2007) has noted the vulnerabilities of oil and gas drilling platforms and refineries along the Gulf of Mexico coast to flooding and high winds associated with extreme weather events. Closure of these facilities and fuel terminals during and after Hurricane Katrina were linked to fuel price increases across the USA. Extreme weather events in non-coastal areas can also affect primary energy supply chains, as we saw in 2008 when heavy snows in central and southern China blocked rail networks and highways used for delivering coal to power plants in these regions. Seventeen of China’s 31 provinces were forced to ration power, affecting hundreds of millions of people in cities across the country (French, 2008). Climate scientists have been wary of attributing these snowstorms to climate change (Perry, 2008), but others note they represent the type of extreme weather event related disruptions that may be more prevalent in the future (Pew Center on Global Climate Change, 2008).
Larsen et al. (2008) note that Arctic transport routes and energy infrastructure critical for moving oil and gas across Alaska are located across areas at high risk of permafrost thaw as temperatures rise. Potential vulnerabilities include structural failure and distribution problems as oil and gas pipelines fracture; reduced access and increased transport costs due to a shorter winter season for ice roads, and increases in repair and maintenance costs. System stresses such as these will produce indirect impacts on cities, generally in the form of higher energy prices.
In developing countries, areas heavily dependent on different types of biomass may be vulnerable to the extent certain climate change risks affect the availability of the material or the transport routes delivering this material to cities. For example, changing temperature levels may reduce biomass availability if plants reach the threshold of their biological heat tolerance or if storms or drought reduce plant or tree growth levels. The extent of these problems will be localized based on how biomass materials are sourced in different urban areas.
Impacts on power generation
There is little within this growing literature, however, that draws specific links between anticipated coastal threats and energy system assets located along or near threatened coastlines. Potential risks exist because many power stations were historically sited along waterways, a legacy of the need for cooling waters that were integral to the design of older thermoelectric power plants. Many facilities also relied on barge deliveries of their coal supply. Given the decades-long lifespan of most large power plants, they now face risks from anticipated sea level rise or more extreme weather events.
Figure 4.5: Location and elevation of power plants along the East River in New York City. Power plant data for 2000 from eGRID (USEPA, 2002) to reflect with recently retired plants deleted. New York City digital elevation model is from the USGS (1999), which has a vertical error of approximately +/-4 feet.
Whether a given power plant (or fuels temporarily stored on-site at or near these plants) is vulnerable to flooding problems is a function of the elevation of the facility, the facility design (e.g., surrounded by berms, etc.), and its proximity to any path that storm-linked tidal surges would follow during extreme weather events. Hurricanes Katrina, Rita, and Ike in the Gulf of Mexico were recent examples where coastal power plants were damaged by storm surges, with several facilities serving New Orleans, Houston, and Galveston forced to shut down due to anticipated flooding. Some remained closed for several days after the storms had passed, suffering extensive wind and water damage (Jovetski, 2006; McKinley, 2008).
Figure 4.5 displays the potential vulnerability of power plants along the East River in New York City to similar storm surges, highlighting the vulnerability of 5,840 MW of power generation capacity at an elevation of less than 5 meters, the height of the storm surge expected in some areas if a Category 3 hurricane directly hits the city.
A different type of risk arises from the fact that cooling waters needed to exhaust waste heat from older thermoelectric power plants may be less able to satisfy their cooling function in the future. Warmer ambient air temperatures and decreased stream flows attributable to climate change increase the risk that power stations will run afoul of rules restricting (ICF, 1995):
• The absolute temperature of water discharged from a power plant
• The absolute temperature of water downstream from power plants and/or
• The temperature rise of waters receiving cooling water effluent from power plants.
Water-cooled power plants are subject to one or more of these standards designed to protect aquatic life, with the exact rules varying by location. To prevent violations, power stations could be forced to scale back their operation or shut down entirely. Such was the case in Europe’s deadly heat wave in 2006, when nuclear power plants in Spain and Germany were temporarily shut or forced to scale back operations due to high receiving water temperatures (Jowit and Espinoza, 2006). Little research has been done to date exploring the overall vulnerability of the energy system to this problem, and doing so – particularly from a city-level perspective – could be difficult, because climate models currently in use cannot be downscaled with pinpoint accuracy to a specific location on a river or bay.
Power plant operations may also be vulnerable to changes in air temperature and air density arising from climate change. The UK Met Office surveyed power plant operators around the UK in 2006, ultimately reporting operator concerns that combined-cycle gas turbines could experience decreased output as temperatures rise and air density decline (Hewer, 2006). Others discount the extent of this problem, however, noting that output reductions will be minor, totaling less than 1 percent under most climate scenarios (Linder et al, 1987; Stern, 1998; Bull et al., 2007), or concluding that it will require systems to be upgraded 1-2 years earlier than otherwise would have been required (Jol-lands et al., 2007). To the extent these problems do occur, they would apply to both central station power plants and district energy facilities. No data have been published thus far exploring such impacts on small-scale (<10 MW) cogeneration systems commonly deployed in many cities as a distributed power generation source.
Although this discussion has thus far focused on traditional thermoelectric power stations, power generation facilities reliant on renewable resources may also be affected by climate change. For instance, hydroelectric facilities fed by glacial and snow melt have historically benefited from the ability of glaciers to regulate and maintain water levels of rivers and streams throughout the summer – a time in many of these regions when precipitation-fed water sources often run low or dry. With increasing temperatures, however, snow levels are decreasing and glaciers are shrinking, jeopardizing the amount of hydroelectric production available to serve many urban areas (CCME, 2003; Markoff and Cullen, 2008; Madnani, 2009).
Changing climate patterns may also affect the timing and level of precipitation available to feed many hydropower systems. For example, although changing precipitation patterns are expected to increase hydropower production by roughly 15-30 percent in northern and eastern Europe by the 2070s, a 20-50 percent decrease in hydropower potential is projected for the Mediterranean region over the same period (Lehner et al., 2005). Problems could arise depending on whether the precipitation falls as rain or snow and at which elevation, because snow serves as a secondary water reservoir, gradually releasing water over the spring and early summer. The elevation at which precipitation occurs is key, because retention dams serve different functions (e.g., water supply, flood control, power generation) based on their elevation and thus have different water release rules. This could affect the availability of power at different times of the year (Linder et al., 1987; Aspen Environmental Group and M Cubed, 2005; Franco, 2005; Vine, 2008).
Whether cities are highly vulnerable to these problems is a function of the type and magnitude of impact of climate change on regional hydrologic conditions and their overall reliance on hydropower. The city of Seattle, Washington, obtains fully 50 percent of its electricity from a network of hydropower dams around the northwestern USA (Seattle City Light, 2005). Projections are that reductions in annual hydropower output in the region are likely by 2080 (Markoff and Cullen, 2008), putting that city’s power supply at risk. Seattle’s municipally owned utility has already begun investing in wind farms on the state border with Oregon to hedge its power generation bets (Seattle City Light, 2005).
Even cities that do not directly rely on hydropower for the electric supply may feel the pinch of declining hydropower availability in their region. Hydropower is generally a low-cost power source, so decreasing availability means it will be replaced by higher cost forms of power, driving up prices as the regional supply market tightens during low-water months or years (Morris et al., 1996).
The impacts of climate change on two other important types of renewable power generation in cities – solar and wind power – are far less definitive. One study examining solar levels in the USA through 2040 projects increased cloud cover resulting from higher CO2 level concentrations could cut solar radiation by 20 percent (Pan et al., 2004). A Nordic study estimates that a 2 percent decline in solar radiation levels could cut solar photovoltaic system output by 6 percent (Fidje and Martinsen, 2006). To the extent cities around the world are seeking to significantly expand deployment levels on local homes and businesses, this could be problematic as long-term power output levels could be less than anticipated. However, because current in-city solar deployment levels are so small compared to overall urban power demand, it will likely be some time before such a decline in solar production becomes significant enough to create major problems. Whether large new concentrating solar facilities recently installed in many parts of the world will suffer degraded output levels is unclear. Several utilities serving urban areas in Spain and the southwestern USA have invested in these projects in rural areas outside of the city (Philibert, 2004), dramatically boosting the level of renewable power feeding the local power system.
Wind patterns (wind speed, duration, and direction) may also change as a result of climate change, although the projected impacts will likely vary seasonally and differ widely from region to region. Research on the Baltic Sea region finds no clear signal on future wind resource levels (Fenger, 2007), while in the UK and Ireland, onshore wind speeds are expected to decrease in the summer and increase in winter (Harrison et al., 2008). Fenger (2007) notes the likelihood that system efficiency levels will increase in Scandinavia during the winter months because of reduced turbine blade icing attributable to warmer temperatures. Research on US wind patterns projects speeds will decline 1 to 15 percent over the next 100 years, depending on which climate models are used (Breslow and Sailor, 2002).
No studies have been identified to date that examine potential wind pattern changes in cities attributable to climate change. Cities are already recognized as being a challenging locale for deploying wind power systems due to the turbulence created by the built environment (Dutton et al. , 2005); the extent to which this may change is unknown and a good area for future research.
Cooling waters and nuclear power in the USA
An August 2007 heat wave forced the shut-down of a reactor unit at the Browns Ferry nuclear power plant in Alabama, USA, leading to international debate about the feasibility of nuclear power in warming temperatures. The Browns Ferry plant uses cooling water drawn from the Tennessee River to condense and cool the steam generated by the plant for its turbines. State environmental regulations impose a 90 °F (32 °C) cap on the river temperature downstream of the plant to minimize stress to aquatic ecosystems, and typically the plant increases the river’s temperature by 5 °F (3 °C). During the heat wave, the upstream river temperature was often at, or above, 90 °F and the plant then became constrained by regulatory limits preventing it from raising the river’s temperature further.
As a result, one unit at Browns Ferry was shut down and power production from another two plants was decreased to reduce the quantity of process steam generated. This allowed the abstracted river water to condense the steam and pass back into the Tennessee River without violating the regulatory limits.
Plant operations were also affected by intake temperatures. In engineering terms, the plant can operate at 100 percent power output with river temperatures up to 95 °F (35 °C), leaving a 5 °F margin between the environmental cap and the engineering threshold. As river temperatures rise, the river water’s ability to condense the steam that drives the turbines drops rapidly, requiring the plant to operate at reduced power outputs. Heat waves increase demand for air conditioning, so Browns Ferry is prone to shut-down when electricity demand is highest.
Climate impacts on energy transmission and distribution
To the extent temperatures are expected to rise as a result of climate change, there may be impacts on the local power grid. In general, transmission and distribution lines and electrical transformers are "rated" to handle a maximum amount of voltage for a fixed period of time before they fail. Changing climatic conditions can lead to such failure by pushing power demand beyond equipment rating levels. In California, for example, a summer 2006 heat wave led to blackouts across the state, as sustained high nighttime temperatures prevented the transformers from cooling down before demand increased again the next morning. Insulation within the transformers burned and circuit breakers tripped, knocking out power for more than one million customers (Miller et al, 2008; Vine, 2008).
Related to the previous discussion on the impacts of rising temperature levels on power plant output is the issue of how temperature changes will affect the power throughput of electricity transmission and distribution lines. When electric current flows through power lines, it encounters resistance from every system component it flows through, which produces heat and results in efficiency losses. These losses normally range from 6 to15 percent of net electricity produced, depending on the age of the system and the degree of electric loading on the lines (EIA, 2009; Lovins et al., 2002; IEC, 2007). The effect on above-ground lines is moderated by the cooler ambient air, while wires below the ground are cooled by moisture in the soil. As temperatures increase, the cooling capacity of the ambient air and soil declines, conductivity declines, and lines may begin to sag or fail altogether (Hewer, 2006; Mansanet-Bataller et al., 2008). Because distributed generation systems tend to involve minimal wiring exposed to the elements, they may be less vulnerable to these problems than central station-based power networks commonly deployed around cities.
Transmission and distribution networks may conversely experience reduced vulnerability as a result of anticipated temperature increases during the winter months, when they are more subject to damage or failure caused by ice or snow storms. Whether cities will be affected by this change will hinge on their general vulnerability to snow and ice storms, the extent to which these patterns change, and their level of reliance on power delivered by these sources. The severe snowstorms plaguing China in 2008 also brought down power lines in many cities and rural areas, compounding the power outages brought about by diminishing fuel stocks (French, 2008).
Electricity transmission and distribution networks may also be vulnerable to storm surges, rising sea levels, and high winds associated with extreme weather events (McKinley, 2008). Whether cities employ above- or below-ground electric wiring systems is largely a legacy of investment or operating decisions made long ago; a blizzard downing wires across the city led local authorities in New York City to move to bury electric wiring in 1888 (New York Times, 1888). Although this tends to eliminate snow and icing problems, it does make the city’s underground transformers and substations more vulnerable to flooding. The local utility in New York is moving to address this problem by installing salt-water submersible transformers in Category 1 flood zones around the city (New York State Department of Public Service, 2007).
Thermal power systems and fuel storage tanks located in buildings may also be vulnerable to sea level rise or storm surges, depending on where they are situated. In many areas, local authorities issue warnings during anticipated flooding events about the need to anchor fuel tanks so they do not shift or flip, spilling their contents and contaminating the building (for example, see State of Maryland (undated)).
