PSE Healthy Energy | May 2020

New Jersey Peaker Power Plants Replacement Opportunities

Across New Jersey, 15 gas- and oil-fired peaker power plants and peaking units at larger plants help meet statewide peak electric demand. These facilities are primarily reliant on combus- tion turbines designed to ramp up quickly and meet . One-third of New Jersey peaker plants primarily burn oil, and two-thirds are over 40 years old; these facilities in particular have high rates of greenhouse gas and criteria pollutant emissions for every unit of generated. Moreover, these plants are located disproportionately in low-income and minority communities, where vulnerable populations al- ready experience high levels of health and en- vironmental burdens. Many of the New Jer- sey peakers operate infrequently, suggesting they may be good targets for replacement with en- ergy storage. The state has also set aggressive © 2020Figure Mapbox © 1:OpenStreetMap Peaker plants across New Jersey clean energy and energy storage deployment tar- gets, providing an opportunity to replace ineffi- The grid in New Jersey is operated by PJM, cient, high-emitting peaker plants in vulnerable which typically defines local requirements for communities throughout the state with energy power capacity on the grid. PJM classifies lo- storage, solar, and other clean cational delivery areas (LDAs), and the EMAAC alternatives. LDA, which covers New Jersey, is considered at least partially transmission-constrained. Local solar and storage deployments may be valuable New Jersey State Policy in these transmission-constrained areas. and Regulatory Environment New Jersey has enacted a suite of policy targets New Jersey Peaker Plants to support clean energy adoption and emission reductions that could facilitate replacement of Peak electricity demand is New Jersey is partially peakers with solar and storage. Key targets in- met by 14 gas turbines and one small internal clude: combustion engine. Features of some of these plants suggest that they should be prioritized for • 2030: Deployment of 2,000 MW of en- replacement with energy storage or a portfolio of ergy storage; intermediary 2021 target of cleaner energy technologies, including: 600 MW. • Aging: Ten are over 40 years old. • 2030: 50 percent of electricity from re- newable resources, including a solar carve- • Short runtimes: Six of the fourteen out and offshore wind targets. plants for which we have data run less than five hours each time they are started up, • 2050: 80 percent reduction in greenhouse which aligns well with battery operation gas emissions below 2006 levels. (see Figure2 ). New Jersey | 2

4

3

2 (MWh)

1

Average Hourly Generation 0 1am 2am 3am 4am 5am 6am 7am 8am 9am 1pm 2pm 3pm 4pm 5pm 6pm 7pm 8pm 9pm 12am 10am 11am 12pm 10pm 11pm

Figure 2: Average hourly generation from the Essex peaker plant. The plant typically meets peak afternoon loads. It ran an average of 6.1 hours each time it started up in 2016, but by 2018 its average run time had declined to 2.6 hours per start, and its declined from 2.6 to 0.2 percent. Batteries can serve a similar role on the grid.

100 tor under 2 percent is 1,400 MW, well below 90 the state’s 2030 storage target of 2,000 MW. New Jersey’s total peaking capacity is 2,732 80 MW. Plants with longer runtimes might be best 70 replaced with a mixed portfolio of cleaner re- 60 sources that can meet similar grid needs, such 50 as solar, storage and demand response.

40

30 Nearby Populations

20 New Jersey peaker plants are located in a mix

Low-Income Population (Percentile) 10 of urban and rural areas, with populations in 0 a three-mile radius ranging from nearly no one 0 10 20 30 40 50 60 70 80 90 (for the peaking unit located at Salem nuclear Minority Population (Percentile) power plant) to 250,000 near the Kearny plant. Primary Fuel Population (3-mile radius) Populations living within three miles of these 0 100,000 200,000 Oil 50,000 150,000 ≥ 250,0.. plants tend to be disproportionately low-income and minority populations: communities near all Figure 3: New Jersey power plants are located but two plants are above the 50th percentile in low-income and largely minority communities. statewide for low-income populations (that is, Bubbles reflect population size. Axes mark state per- they have more low-income households than half centiles for low-income (double federal poverty limit) and minority populations living within three miles of of New Jersey census tracts), and communities each facility. near nine plants are above the 50th percentile for minority populations (see Figure3 ). Many communities also have a high cumulative expo- • Infrequently used: 11 operate at a ca- sure to environmental health burdens from nu- pacity factor of 2 percent or less—that is, merous sources. We developed a cumulative vul- they generate 2 percent of the electricity nerability index that integrates data on health that they would if they were running con- burdens (asthma, heart attacks, premature birth stantly at full power year-round. The five rates); environmental burdens (ozone, particu- oil-fired peakers all operate at a capacity late matter, toxics, traffic proximity, lead paint, factor of 0.6 percent or less. and hazardous facilities); and demographic indi- cators (low-income, minority, linguistically iso- The net capacity of facilities with a capacity fac- New Jersey | 3

225

200

175

150

125

100

75

50 Cumulative Vulnerability Index

25 0 Essex Kearny Station Sayreville Mickleton Cumberland Forked River West Station Carll's Corner Sherman Avenue Ocean Peaking Power Burlington Generating Burlington Salem ( unit) Linden (gas turbine unit) Gilbert (gas turbine unit) turbine (gas Gilbert Seaside Heights Power Plant EJ Indicator Category Health indicator Environmental burden indicator Demographic indicator

Figure 4: The cumulative vulnerability index reflects a set of environmental, human health and de- mographic indicators for populations living within three miles of each plant. The score is based on a comparison of indicators to statewide values: if a plant ranked at the median percentile for all indicators, it would score 150, which is indicated by the red dashed line. (Note: Salem has limited data available.) lated, and non-high school educated popula- Summary tions). The cumulative vulnerability index for populations living within three miles of each fa- New Jersey peak demand is met by an aging cility is shown in Figure4 . fleet of peaker power plants located dispropor- tionately in the state’s low-income and minority communities. The state’s oil-burning plants, in Emissions and the Environment particular, are used infrequently but have high pollutant emission rates when they are operated, One-third of New Jersey peaker plants and units suggesting they might be good candidates for burn primarily oil and the remainder primarily replacement. The state’s energy storage targets use natural gas, although many burn both. The provide an opportunity to target the more ineffi- oil-burning facilities in particular, as well as a cient and polluting facilities, particularly in dis- couple older natural gas turbines, have high ni- advantaged communities, for replacement with trogen oxide (NOx) emission rates—pollution cleaner alternatives. In the attached table, we per unit of electricity generated. NOx is a pre- provide operational, environmental and demo- cursor to ozone and particulate matter forma- graphic data for New Jersey peakers and nearby tion. The state is considered out of attainment populations. Indicators such as nearby popu- of federal ozone standards; operation of these lation, emission rates, heat rate (fuel used per plants on hot summer days to meet air condi- megawatt-hour), operation on poor air quality tioning demands can exacerbate these poor air days, capacity factor, and typical run hours can quality conditions. Notably, between 2016 and also inform whether a given plant might be a 2018, the Ocean Peaking Power plant generated good target for replacement with storage, so- 10 percent of its electricity on days exceeding lar+storage, demand response, or other clean local air quality standards. alternatives. These data should be accompa- nied by engagement with affected communities to determine replacement priorities and strate- gies. New Jersey | 4 13 105 214 202 183 237 CVI 12 (51) (65) (87) (80) (80) 19% 28% 51% 42% 42% tile) % low- income (percen- 11 (8) 6% (60) (74) (53) (76) 48% 69% 39% 73% white tile) % non- (percen- Pop. 10 % high MWh ozone days 9 x NO rate lbs/MWh 2 8 CO rate tons/ MWh 7 Heat rate MWh MMBtu/ 6 Run hours/ start 5 factor Capacity 4 53 2.0% 3.830 9.8 0.6 8.2% 7.8 1.0 7.9% 9.0 75,794 0.5 0.3 1.0% 4,102 47 1.8% 5.849 24.331 1.2% 1.5 5.1 17.8 1.4% 10.0 3.9% 6.2 25,279 0.6 13.8 0.8 0.8 3.2% 2.2 221,376 7.9% 23,741 Age contributes to ozone and particulate matter formation. 3 x Load zone PSEG AECO AECO PSEG JCPL For methods see: www.psehealthyenergy.org. 2 84 94 77 242 231 MW 1 Natural gas Natural gas Natural gas Natural gas Natural gas New Jersey peaker plant operational and demographic data. Deerfield River ) emitted per unit of electricity generated; NO x Plant descriptionStatus City Fuel Operation and emissions Demographics (3-mile radius) Operating Burlington OperatingOperating Upper Millville Operating Forked (2399) (EIA ID) (2401) Operating Newark Name Burlington Generating Station Carll’s Corner (2379) Cumberland (5083) Essex Forked River (7138) Percent of generation onPercentile days minority nearby population monitors indicates recordPercentile percent exceedances low-income of of population census federal indicates tracts ozoneCumulative percent across Vulnerability standards. of Index the census combines state tracts state with across percentiles lower the for fraction state demographic, of with health non-white lower and populations. fraction environmental of exposure households indicators. below A double median the on federal all poverty values limit. would score 150. Primary fuel; many plantsInstalled burn nameplate both capacity oil (plant andLoad size). natural zones gas. within PJM,Age indicating of utility oldest service unit area. Percent in of 2020. time runningAverage as number compared of to hours runningHeat plant all rates runs year are each at energy timeDirect full burned it carbon capacity. per is dioxide unit turned emissions of on. Nitrogen per electricity oxides unit generated; (NO of high electricity heat generated; rates does reflect not low include efficiency. upstream emissions. 1 2 3 4 5 6 7 8 9 10 11 12 13 New Jersey | 5 94 76 179 232 224 167 153 206 199 (55) (58) (78) (77) (57) (80) (86) (84) (28) 21% 23% 40% 39% 22% 42% 49% 46% 17% (57) (18) (79) (77) (34) (37) (70) (73) 44% 78% 74% 22% 25% 63% 67% NA% 8 0.6% — 11.9 1.0 26.6 — 36.7 48 0.2% 3.4 17.9 1.2 2.7 0.1% 55,684 53 1.5%46 4.9 1.0%17 12.3 6.2 10.1%49 0.8 26.0 13 0.02% 0.6 1.5 10.1 1.3 8.6% 28.8 155,967 0.6 30.9 6.1% 2.5 0.3 20,393 9.7% 37.129 83,116 0% 4.7%48 — 7.8 0.3% 12.8 3.6 — 0.8 16.2 — 0.9 1.3 — 4.4% 6.4 22,610 2.1% 38,848 53 7.1% 6.0 10.0 0.6 0.3 3.9% 251,073 50 0.4% 4.7 11.6 0.8 1.4 0% 5,254 3% (3) — JCPL PSEG AECO JCPL PSEG AECO — PSEG JCPL 6 71 27 212 384 383 383 113 605 161 Oil Natural gas Natural gas Natural gas Natural gas Oil Natural gas Heights Alloways Creek Township Operating Seaside Operating Sayreville Oil OperatingOperating Linden Operating Mickleton Operating Lakewood Lower OperatingOperating Vineland Vineland Oil Operating Milford Oil 14 15 16 (7288) (2404) Operating Kearny GT GT (55938) GT Seaside Heights Power Plant (58172) Sayreville (2390) Linden (2406) Mickleton (8008) Ocean Peaking Power Salem (2410) Sherman Avenue West Station (6776) Kearny Gilbert (2393) Gas turbine unit atGas 512 turbine MW unit gas at combinedGas 1,062 cycle turbine MW plant. unit gas at combined 2,330 cycle MW plant. plant. 14 15 16