NORTH AMERICAN POWER PLANT AIR EMISSIONS

Facilities in Canada are required to report their emissions of CACs to the NPRI if the emissions of a given contaminant exceed the specified reporting threshold. Reporting threshold levels for CACs were issued in 2002 and were still in effect in 2005. The reporting thresholds were as follows: 20 tonnes (air release-based) for CO, NOX, SO2 and Total PM; 10 tonnes for VOCs; and 0.5 and 0.3 tonnes for PM10 and PM2.5, respectively. These reporting requirements were expected to cover 90% of all facilities [26]. The threshold for mercury and its compounds was set at 5 kg and applied not only to emissions, but also to manufacturing, processing or otherwise using the substance.

Emissions data for greenhouse gases (GHGs) are not included in the NPRI. Facility-specific GHG emissions data are gathered through the Greenhouse Gas Emissions Reporting Program (GHGRP) [27], which was established in Canada in 2004 and required reporting from all facilities emitting more than 100 kt CO2-eq of GHGs annually (in 2009, the GHGRP reporting threshold was lowered to 50 kt CO2-eq [28]).

Another important source of information is Statistics Canada. This government agency issues periodic reports on Electric Power Generation, Transmission and Distribution that contain valuable information about technologies used, installed capacity and the geographical distribution of generating stations. From one of these reports [29] total provincial and national values were obtained relative to electricity generation, fuel utilization, and greenhouse gas emissions.

Plant-level annual electricity generation data were obtained for several facilities directly from their parent companies’ annual reports, environmental performance reports or other publicly available corporate documents (see Annex for further references).

Electricity in Canada is also produced by cogeneration. Thus, electricity can be produced in plants that are not exclusively dedicated to this purpose. In 2005, there were numerous facilities taking advantage of the higher efficiencies obtained via processes such as combined heat and power to produce the heat and electricity they require and even put some back into the grid. Therefore, a number of electricity producers are classified in industry categories other than the NAICS code 221112 corresponding to “fossil fuel electric power generation.” However, for this report, only the 189 facilities classified under that code were considered.

Annual electricity generation figures for some Canadian power plants were not published. In those cases, the electricity generation was estimated on the basis of the CO2 emissions, typical heat rates for the technology involved, and taking into account the overall generation by province. Further details on the estimation methodology and specific data sources can be found in the Annex.

With the variety of independent sources of information, it was expected that each might yield different data, which was indeed the case. For example, some of the facilities that reported to the NPRI did not report to the GHGRP and vice versa, and some of the facilities listed in the Statistics Canada database do not report to the NPRI or the GHGRP. Nevertheless, with the information obtained from the sources previously mentioned, a single, unified database was created using the facility IDs included in the GHGRP and the NPRI databases.

This database includes the following parameters: industry sector (NAICS) code, location, fuels and technology used, generating unit details, installed capacity, emissions, annual electricity generation, and pollution control data.

Information Sources and Methodology for Electricity Generation Data of Canadian Power Plants

Saskatchewan
For the Saskpower plants that use lignite as fuel, the estimated electricity generation corresponds (within 1.5%) to the overall net generation reported by the company [1]. Heating values used for the calculation were the national averages by type of fuel [2].

Newfoundland and Labrador
The electricity generation values for the Province of Newfoundland and Labrador [2] were matched to the generation of the only coal-fired plant in that province.

Nova Scotia
Emission rates for the Nova Scotia Power (NSP) coal-fired power plants were obtained from the NSP web site [3]. The Nova Scotia Power emission rates are disclosed under the Carbon Disclosure Project, an international, independent, not-for-profit organization that maintains a database of climate change information. The total generation derived from these emission rates and the NPRI data for CO2 emissions agreed to within 2.7% with the total generation reported for the corresponding technology in that province [2]. The three NSP combustion turbine generating stations were considered together since a single datum was found for CO2 emissions from the three facilities [3]. The corresponding electricity generation for these three facilities was found to be negligible in comparison with the generation from this type of technology reported for that province [2].

New Brunswick
In the province of New Brunswick, NB Power Generation Corporation operated 5 steam power plants in 2005: Dalhousie, Belledune, Grand Lake, Courtenay Bay, and Coleson Cove. In addition, there were three combustion turbine plants (Ste Rose, Millibank, Grand Manan) for which information was not available from the NPRI nor the GHGRP. However, according to the 2005 NB Power (NBP) annual report [4], the first two belong to the Environmental Management Systems (EMS) of Belledune, and the last one belongs to the EMS of Coleson Cove, so it is possible that emissions from these combustion turbine plants are reported as part of those larger power plants. Courtenay Bay Generating Station (CBGS) had three units, but in 2002 Irving Paper Ltd assumed responsibility for the operation of Unit 2 (NPRI ID No. 8003). Units 1 and 3 were still the responsibility of NBP (NPRI ID No. 1706) but NBP did not operate CBGS in 2005 [4]. Emissions of CO2 were reported to the GHGRP for CBGS unit 2, under NAICS code 221330, even though the facility was registered for the NPRI with a NAICS code 221112.

The emissions of CO2 obtained from the GHGRP for the five steam power plants were consistent with the overall system fossil fuel generation emission rates [5] and the total electricity generation reported by Statistics Canada [2]. The estimated electricity generation for each individual plant based on the CO2 emissions was also consistent with the overall generation reported by Statistics Canada [2]. One additional combustion turbine plant, the Bayside Power Plant (then owned by Irving Oil Power L.P.), accounted for the electricity generation reported by Statistics Canada [2] for this technology. It has a combined cycle natural gas turbine [6] with an estimated net thermal efficiency of 51.6%. Efficiencies of this order and even higher have been reported for this type of technology [7, 8, 9].

Ontario
Ontario Power Generation (OPG) owned and operated 6 fossil-fueled power plants with steam turbine generators in 2005. The Lakeview Power Station operated only during the first third of the year, after which it was shut down permanently [10]. The electricity generation of this power plant was estimated on the basis of its CO2 emissions and an assumed thermal efficiency typical of coal power plants [11]. For the remaining five plants, the net generation was obtained from an OPG public report [12]. The Fort Frances plant, owned by Abitibi Consolidated Inc., also operated during 2005. Its generation was estimated from its reported CO2 emissions using an efficiency [rating] similar to the other steam generating power plant in Ontario, the EPCOR Tunis power plant. The generation for the Tunis plant was estimated from the Ontario EPCOR operations total generation [13], which was then apportioned to each plant of the Ontario operations according to the installed capacity of each plant.

The total generation during 2005 from the steam generating fossil-fueled plants in Ontario considered in this report differed by 3.2% from the figure indicated in the Statistics Canada report [2]. However, combustion turbine generating plants in the province consisted of two types, single and combined cycle, therefore different thermal efficiencies were assigned to them in order to estimate the generation from CO2 emissions. There were also some plants that had units of both types, that is, steam and combustion turbines, but it was not possible to estimate the generation for each type, as CO2 emissions are reported for the whole plant. Therefore, the value estimated for the total generation from steam turbines is slightly smaller than the value reported by Statistics Canada [2]. The difference between the total estimated and reported generation is less than 0.1%.

Québec
Emissions of CO2 from two plants in Quebec, Tracy and Cap-aux-Meules, were reported in the GHGRP. The former is a 600 MW steam generating heavy fuel oil-fired plant dating back to the 1960s [14, 15]. In 2005, this was only operated when the system was at peak demand. Old, low-efficiency steam plants are generally used for peaking [16]. The Cap-aux-Meules plant is one of the province's diesel-fueled internal combustion plants. Hydro Quebec published the total generation for its four main fossil fuel generating plants as well as their corresponding CO2 emissions [17, 18]. It was possible to determine the combined generation for Tracy and the other three plants, Bécancour, La Citière and Cadillac. These three are light fuel oil-fired, using combustion turbines, and are also used for peaking. Therefore, it was reasonable to treat them as a unit.

The total emissions from these four power plants can be expressed as:

E_T = r₁G₁ + r₂G₂

where

E is the CO2 emissions (Mg)

r is the CO2 emission rate (Mg/MW-h)

G is generation in MW-h

the subscript 1 refers to Tracy, the subscript 2 refers to the three combustion turbine plants together, and the subscript T refers to the total of the 4 plants considered.

Obviously,

G_T = G₁ + G₂

and

E_T = E₁ + E₂

where

E₁ = r₁G₁

and

E₂ = r₂G₂

Accordingly, the generation from Tracy can be determined as a function of the total generation, the total CO₂ emissions of the four plants and those from Tracy (all of which are known) along with the emission rate of Tracy.

where

and the following relation holds between the emission rates

The emission rate at Tracy (r1) was estimated using the following relationship:

Where F is the CO₂ emission factor for steam generating plants using heavy fuel oil, assumed 3.124 kg/L [19]; H is the net heat content of heavy fuel oil, assumed 36,813.47 kJ/L [20]; and η, the thermal efficiency, assumed 28%. This efficiency value was chosen considering the lowest limit of the range of efficiencies observed for this type of plants. In 2000 there were already concerns about the high emission rates and low efficiency of this plant [15]. Also, since the plant was operated at low capacity factors, the efficiency could have been reduced even further [15, 21].

By following the above procedure, all the reported values for both generation and emissions are made consistent.

In 2005, Hydro Quebec operated one steam generating plant (Tracy), three light fuel oil combustion turbine plants (Bécancour, La Citière and Cadillac) and 24 diesel-fueled internal combustion plants for a total of 1595 MW of installed capacity [18]. The internal combustion plants provided off-grid base-load electricity, mainly in the northern regions. The total electricity generated by these 24 plants in 2005 was 277.1 GW-h [22, 23], although the plant at Cap-aux-Meules generated nearly 65% of the total. This plant also reported emissions of CO₂ to the GHGRP.

The Boralex generating station at Kinsey Falls was registered in the GHGRP under NAICS code 221112 ("fossil fuel electric power generation"), whereas in the NPRI it was registered under NAICS code 221119 ("other electric power generation"). Since emissions of CO₂ exceeded the threshold value for reporting and the plant used natural gas, it was included in this report with the NAICS code for fossil fuel electricity generation. The generation of the Kingsey Falls plant was estimated on the base of its CO₂ emissions.

Alberta
Generation data for Alberta were obtained from ATCO Power [24] for its plants in that province, and from the Alberta Electric System Operator (AESO)'s web page [25]. Generation data for some of the plants indicated in the table were not publicly available and the value was therefore estimated on the basis of the CO₂ emissions and efficiencies of similar plants.

British Columbia
There are inconsistent values reported for British Columbia. For example, the Burrard Thermal Generating Station is a conventional natural gas-fired generating station that generated 456 GW-h [26]. The heat rate of this plant for years 1999 and 2000, based on generation and energy input reported in a cost benefit evaluation of the plant made in 2001 [27], was found to be approximately 10,000 Btu/kW-h, which is standard for this type of plant. Using this heat rate, the quoted generation, and either typical CO₂ emission factors [28] or the reported CO₂ emission intensity [27], a CO₂ emission of between 239.4 and 244.8 kt of CO₂ is obtained. The value reported to the GHGRP for the CO₂ emissions from Burrard in 2005 is only 68.1. This value is obviously inconsistent with the generation reported by BC Hydro. From an analysis of the CO₂ emissions reported to the GHGRP and the generation reported by BC Hydro, it is clear that the data inconsistency spans several years. It seems that the GHG values reported by BC Hydro are adjusted by overall GHG reduction actions [29].

The generation of the other steam turbine plant, Duke Energy Gas Transmission - McMahon Cogeneration Plant, that operated in BC in 2005 was obtained by adjusting the thermal efficiency so that the total generation from steam turbine plants matched the value found reported [2]. Finally, the two combustion turbine plants considered were probably those with the highest generation among the plants in BC using this technology. A number of internal combustion plants operated in BC during 2005 for back-up or peak supply purposes, but these were not considered here, as no information was available. The total relative difference for each technology was less than 1%.

Previous