Estimating Power Plant Contributions to Surface Pollution in a Wintertime Arctic Environment

Abstract

Arctic winter meteorology and orography in the Fairbanks North Star Borough (FNSB, interior Alaska) promote stably stratified boundary layers, often causing acute pollution episodes that exceed the US-EPA National Ambient Air Quality Standards. Power plant emission contributions to breathing level (0-10 m) pollution are estimated over the FNSB using high-resolution Lagrangian tracer simulations run with temporally varying emissions and power plant plume rise accounting for atmospheric boundary layer stability and validated against comprehensive ALPACA-2022 observations. Average relative power plant contributions of 5-23% and 4-28% are diagnosed for SO₂ and NO _x , respectively, with lower relative contributions in polluted conditions due to larger surface emissions. Highest population-weighted contributions are found in central and eastern (residential) areas of Fairbanks. Significant temporal variability in power plant contributions is revealed, depending on power plant operations and Arctic boundary layer stability. Vertical transport of power plant tracers to the surface depends on the interplay between the presence of temperature inversion layers and power plant stack heights as well as prevailing large-scale or local winds. Notably, power plant emissions can be transported to the surface even under strongly stable conditions, especially from shorter stacks, whereas down mixing from tall stacks mainly occurs under weakly stable conditions.

Figure 1

Hourly averaged surface (gray) and power plant emissions (colors) emitted in the FNSB area (kg hour^–1) each day for (a) SO₂ and (b) NO_x according to the ADEC emissions inventory and the emissions from power plants during the ALPACA-2022 campaign. The black stars indicate weekend days on panels a and b. (c) ADEC emission sector contributions (kg hour^–1) for the total surface emissions for NO_x (left bars) and SO₂ (right bars). (d) Left axis: observed hourly PM_2.5 (μg m^–3) at the NCore measurement site (Downtown, black line). The ‘polluted’ conditions, used in Section 3, for PM_2.5 are shown by the thick red lines above. Right axis: observed hourly SO₂ (parts per billion, ppb) at the CTC measurement site, colored by SS (green) and WS (orange) stability regimes as defined in Brett et al.

Figure 2

Map of Fairbanks and North Pole. Solid and dashed lines indicate the FNSB and Fairbanks EPA nonattainment areas, respectively. The power plant locations (white triangles) correspond to the following power plants: (a) Aurora, (b) Zehnder, (c) University Alaska Fairbanks (UAF), (d) Doyon (Fort Wainwright), (e) North Pole. Grid cells (1.33 km resolution) with available population counts are shown. Analysis areas are depicted with the white borders (see text for details). OpenStreetMap contributors 2024. Distributed under the Open Data Commons Open Database License (ODbL) v1.0.

Figure 3

(a) Total δSO₂ (ppb) power plant enhancements (ppb) at 0–10 m (campaign average). (b) Total SO₂ power plant population-weighted contributions (PWCs), see Section S1.3 for details. The measurement sites are indicated by the colored diamonds and power plants shown by the white triangles, as in Figure 2.

Figure 4

Total power plant δSO₂ (ppb) enhancements (black lines) between 0 and 10 m as a function of time (Alaskan Standard Time, AKST), colored by the different power plant contributions, indicated in the legend. Panels (a) to (f) correspond to the different analysis areas, indicated in Figure 2. The gray shaded periods correspond to SS conditions, and the nonshaded periods correspond to WS conditions. See text for details.

Figure 5

Cases illustrating downward power plant plume transport on (a) 23 January and (b) 30 to 31 January at the CTC site (situated Downtown, see Figure 2), showing (i) log(RCS) of wind LiDAR data as a function of altitude (40–290 m) and time, indicating power plant plumes aloft, (ii) total simulated power plant δSO₂ (ppb) as a function of altitude (0–290 m) and time (Downtown area), and (iii) total simulated power plant (PP) δSO₂ (ppb) at 0–10 m (Downtown area), for the stacks indicated (total = black line). The red dotted line corresponds to the 3–23 m δCO₂ (parts per million, ppm) observed at CTC, see text for details.

Figure 6

Hourly absolute power plant contributions (ppb) at breathing level for SS vs WS regimes averaged over each area for tall power plant stack heights (>30 m, left panels) and short power plant stack heights (<30 m, right panels) for (a) δSO₂ and (b) δNO_x. Box lower edge = 25th percentile, upper edge = 75th percentile, and middle line = 50th percentile (median). Lower whisker = lowest data point within 25th percentile minus the interquartile range (IQR) × 2, and upper whisker = 75th percentile plus IQR × 2. Scatter points are outliers.

Figure 7

(a) δSO₂ power plant concentrations (ppb) between 0 and 10 m for data when power plant concentrations >0.1 ppb. (b) δSO₂ power plant contributions relative to the total δSO₂ tracer (surface + power plant) in %. (c) SO₂ population-weighted contributions (PWC) (see Section S1.3). Average values over the campaign (solid lines) and during ‘polluted periods’ (dashed lines) are shown. Results are shown for each area defined in Figure 2, and minimum and maximum values correspond to sensitivity simulations, see text for details regarding the upper and lower limits of the bars.