Chapter 3. Atmospheric iron deposition in the Northwestern Pacific Ocean and its
3.5 Conclusion
Due to the rapid economic development in East Asia, anthropogenic aerosols have
become an important Fe source in the surrounding oceans. This study applied a regional
meteorological and air quality model to simulate the deposition of atmospheric iron in
the ECS, SCS, and the NWPO. The quantity of fly ash iron emission was estimated by
linking them to SO2emissions, using a composition factor based on the iron and sulfur
content of coal and typical fly ash, as well as a recovery factor that takes into account
the removal of fly ash and sulfur before waste gases are emitted into the atmosphere.
Special consideration was given to iron emissions from steel and iron plants, in which
iron content is substantially higher than in the fly ash typically encountered in coal
burning.
Simulations revealed large amounts of iron being deposition in the NWPO and
NSCS from both mineral dust and fly ash. The industrial coal burning was the largest
source of iron associated with fly ash, among which steel/iron plants were the largest
contributors, accounting for more than 50% and 60% of the fly ash iron deposited in the
NWPO and NSCS, respectively. Iron from mineral dust remains the main source of iron
over the NSCS and NWPO, contributing 85% and 92% of the total iron input,
respectively. However, when taking into account the high solubility of
combustion-related iron, the proportion of bioavailable iron from anthropogenic sources
may well exceed that from natural sources. Over 77% and 87% of the dust iron entered
NWPO and NSCS, respectively, via wet deposition. However, dry deposition played a
more important role in fly ash iron, accounting for 50% of the deposition in the NWPO
and 79% in the NSCS.
Our model results indicate that seasonal variations in iron deposition over the
NSCS and NWPO were mainly due to the changes in atmospheric circulation system
(such as monsoon), and cloud and precipitation systems. Large seasonal variations were
observed for Fe deposition fluxes and the relative contributions of fly ash and mineral
dust. Over the NWPO, the highest deposition rate of dust iron occurred in the spring,
whereas the highest deposition rate of fly ash iron was in the winter. Over the NSCS,
the strongest deposition of dust iron as well as fly ash iron occurred in the autumn. The
lowest rate of iron deposition occurred in the summer, and the values are 5 times lower
than the maximum seasonal rates for both oceans. The influence of human activity was
most pronounced in the summer, at which time contributing 14% and 28% of the total
iron was deposited into the NWPO and NSCS, respectively.
This study has demonstrated a strong contribution of iron deposition from
anthropogenic sources in the Northwestern Pacific Ocean and the marginal seas of East
Asia using year 2007 as an example. It would also be interesting to examine how trends
in coal use and the application of emission scrubbers in surrounding countries affect the
anthropogenic nutrient input. Also, the importance of the Fe input associated with
anthropogenic aerosol deposition in oceanic biogeochemistry deserves further
investigation.
Table 3.1: Modal size (Pm) and spectral width of dust size distribution used in the model. Spectral width is defined as the standard deviation of the lognormal distribution.
Land type fine mode coarse mode
Modal size Spectral width Modal size Spectral width
Desert 0.07 1.80 1.19 2.31
Loess Plateau 0.07 1.78 0.83 2.52
Urban 0.07 1.80 1.00 1.80
Others* 0.07 1.79 1.16 2.38
*Others include forest, cropland, and grasslands, etc.
Table 3.2: Simulated seasonal variations in iron deposition (in Pg m-2day-1) in the NWPO and NSCS for the year 2007. Values in the parentheses are the fraction (in %) of total deposition attributable to wet deposition.
Table 3.3: Simulated seasonal variations in iron deposition (in Pg m-2d-1) and chlorophyll a concentration (in Pg m-3) in the NWPO and NSCS in 2007. Total iron is the sum of dust iron and fly ash iron; “Soluble iron (4%)” and “Soluble iron (40%)” indicate fly ash iron with 4% and 40% solubility, respectively; dust iron solubility is fixed at 0.45%., Values in the parentheses are the percentage of fly ash iron in total iron or climatological (2001-2010) mean Chl a concentrations. The highest seasonal values are highlighted in underlined bold font, and the second highest values are in bold font.
Spring Summer Autumn Winter Annual
NWPO
Total iron 153(5) 28 (15) 65 (9) 140(8) 96 (7)
Soluble iron (4%) 1.0(31) 0.3 (60) 0.5 (47) 1.0(42) 0.7 (41)
Soluble iron (40%) 3.6(82) 1.8 (94) 2.6 (90) 4.8(88) 3.2 (87)
Chlorophyll a 87(78) 47 (51) 49 (64) 85(91) 67 (71)
NSCS
Total iron 103 (13) 36 (28) 221(14) 168(15) 132 (15)
Soluble iron (4%) 1.0 (58) 0.5 (78) 2.1(59) 1.6(60) 1.3 (61)
Soluble iron (40%) 5.9 (93) 4.1 (97) 13.1(93) 10.5(94) 8.4 (94)
Chlorophyll a 110 (110) 121 (123) 241(150) 205(156) 170 (135)
Figure 3.1: Domain of model simulation (light blue shading), and the area of the ocean that was analyzed (dotted rectangles). Red dots indicate the location of observation sites. D: Dongsha Island; S: South East Asia Time Series Station; P: Pengjia Islet; C: Cape Fuguei; W: Wanli; X: Xiaomen Islet; M: Magong.
Figure 3.2: Mass fractions of Fe and Al in various types of dust and soil particles, coal, and combustion ashes. Red circles and black dots enclosed by red circle represent typical fly ashaand bottom ashd, respectively; red dots represent fly ash from steel plantb; green dots with red circles represent fly ash from biomass burningc; black squares represents coale; orange, blue, green and purple diamonds represent desert dustf, river dustg, agriculture and topsoil dusth, and urban dusti, respectively. The blue-shaded area represents the range (i.e., 1.3 to 2.7) of Fe to Al ratios in particles collected in seawater from the SEATS 2007 observations [Ho et al., 2010]. Sources pertaining to the data points are listed in Tables B and below.
a: Yuan et al. [1998]; Liu et al. [2004]; Koukouzas et al. [2006]; Liu et al. [2007]; Cao et al. [2008];
Wang et al. [2008]; Vassilev and Vassileva [2009]; Zhang et al. [2010]; Tang et al. [2012]; Zhang et al.
[2012]. b: Prati et al. [2000]; [2007]; Tsai et al. [2007]. c: Gaudichet et al. [1995]; Yamasoe et al. [2000].
d: Liu et al. [2004]; Wang et al. [2008]. e: Querol et al. [1997]; Liu et al. [2001]; Zhang et al. [2004]; Dai et al. [2005a]; Dai et al. [2005b]; Dai et al. [2006a]; Dai et al. [2006b]; Yang [2006]; Song et al. [2007];
Wang et al. [2008]; Yang [2008]; Sun et al. [2010]; Zhou et al. [2010]; Li et al. [2012]. f: Li et al. [1984];
Tian et al. [1993]; Water Research Institute of Nagoya University [1991]; Mori et al. [2003]; Zhang et al.
[2003]. g: Li et al. [1984]; Tan et al. [2006]; Lin [2010]. h: Su et al. [2007]; Li et al. [2009]; Yanai et al.
[2012]. i: Dong et al. [1984].
Figure 3.3: Mass fractions of S and Fe in various types of coal and combustion ashes. Red circles represent fly ash, dark dots with red circles represent bottom ash, and black squares represent coal. The black line is the geometric mean of S:Fe (0.62) for coal, and the red line is the geometric mean of fly ash (0.086); the line for the geometric mean of bottom ash (0.088) overlaps with the red line. Sources of the data points are listed in Tables 3.A1─3.A3.
0.01 0.1 1 10
0.1 1 10 100
S (% )
Fe (%)
CoalFly ash Bottom Ash
Figure 3.4: Simulated mean emission flux (in mg m-2d-1) of mineral dust (top) and fly ash (bottom) over East Asia in 2007. Note the difference in color scales.
Fly Ash Emission Dust Emission
Figure 3.5: Simulated seasonal mean concentration (in μg m-3) of near-surface mineral dust in 2007.
Spring Summer
Autumn Winter
Figure 3.6: Simulated seasonal mean concentration (in μg m-3) of near-surface fly ash in 2007.
Spring Summer
Autumn Winter
Figure 3.7: Monthly mean mass concentration of particulate matters observed and simulated at the Magong (left) and Wanli (right) stations in 2007. Black lines (TEPA): TEPA measurement of PM10; black dashed lines (Total): total simulated PM10; orange dash lines (TAON): mineral dust PM10estimated from TAON measurements; orange dash lines (Dust): simulated mineral dust PM10; red dash lines (FA):
simulated fly ash PM10; blue dash lines (Other): all other simulated PM10.
Figure 3.8: Simulated seasonal mean dust iron deposition (in μg m-2d-1) for the year 2007.
Spring Summer
Autumn Winter
Figure 3.9: Simulated seasonal mean fly ash iron deposition (in μg m-2d-1) for the year 2007.
Spring Summer
Autumn Winter
Figure 3.10: (a) Fraction (in %) of fly ash iron deposition in total iron deposition; (b) Fraction (in %) of fly ash soluble iron deposition in total soluble iron deposition; (c) Fraction (in %) of combustion iron deposition associated with steel plant emissions. All fractions were calculated for the year 2007.
(a) (b) (c)
Figure 3.11: Simulated percentage of iron deposition from fly ash in total deposition over the NWPO (top panel) and NSCS (bottom panel) with fly ash composition factor D (columns) and recovery factor E (symbols) of different values. Columns from left to right (dark to light shading) are values calculated usingD = 6.06 (upper 1 geometric standard deviation), 1.5 (mean) and 0.04 (lower 1 geometric standard deviation), respectively, and withE for the year 2007. Symbols represent values calculated using various E values at D = 1.5: “▲” for E2001, “” for E2007and “▼” for E2009.
Percentage of Fe Deposition from Fly Ash
0
Percentage of Fe Deposition from Fly Ash
0
Figure 3.12: Simulation results for Fe:Al ratio in 2007 compared with those observed at the following locations and times: the Pengjia Islet (green) in 2005cand 2010c(unpublished data); East China Sea (red) in 2007band 2008c; Dongsha Islet (blue) in 2007aand 2010c. Also included are the seawater samples from the SEATS station (purple) collected in 2007a. The superscripts indicate the data source: a) Ho et al.
[2010]; b) Hsu et al. [2010]; c) unpublished data. The square, circle, triangle, and diamond symbols represent seasonal averages in spring, summer, autumn and winter, respectively. These seasonal averages were calculated by taking the mass-weighted geometric mean of daily data. The cross bars indicate the range of seasonal variation in both simulation and observation data at one geometric standard deviation, and line with different styles indicate observations obtained during different years. Model results were interpolated to the location of the observation sites, with the exception of shipboard measurements obtained in the East China Sea for which the corresponding model results are area averages.
0.3