Nitrogen oxides

NO, NO² , PAN, HNO²

Modelling concentrations of NO2 using the photo-stationary equilibrium reaction requires estimates of O3

(background) concentrations on a local scale. Such O3 concentrations are strongly influenced by neigh-bouring NO sources, making this approach unsuited for this model. Basically, the OPS model computes contributions of sources independent of each other, so empirical relations between NO and NO2 concen-trations cannot be used unless the ‘background’ NO2 concentration is taken into account. An alternative would be an iterative approach, i.e. first calculating total concentrations linearly and then the non-linear relations using the results of the first step as the background levels. The computed NO2,PAN and HNO2

concentrations would not be very accurate anyway. These considerations have led to the choice of model-ling the sum of NO, NO2, PAN and HNO2 as a single conservative species NOx.

Measurements in Delft carried out by TNO indicate a PAN concentration which, on average, is only in the order of 5% of the NO2 concentration (Ogilvie, 1982). The deposition properties are also uncertain but probably not very different from those of NO2; it was therefore decided not to take PAN into account as a separate component for this model but to consider it as a part of NOx.

Slanina et al. (1990) report average nitrous acid (HNO2) concentrations of 0.64 ppb for a forest site in the Netherlands (Speulderbos), which is in the order of 4 % of NO+NO2 concentrations. Similar results are reported by Kitto and Harrison (1992).

In OPS, modelled 'NOx' consist of NO + NO2 + HNO2 + PAN, including a 4% contribution of HNO2 and a possible PAN contribution of 5% to the NO2 concentration. However, measurements of NOx usually consist only of NO + NO2. At an average NO2/NOx ratio of 0.65, the modelled NOx (NO + NO2 + HNO2

+ PAN) concentration may be systematically 8% higher than measured NOx (NO + NO2) concentrations.

In order to keep the model output consistent to reported concentrations, an 8% reduction is applied to the model output of NOx concentrations, such that the model output of NOx represents the sum of NO and NO2 .

Nitrate formation in EMEP

An important source of aerosol nitrate in the troposphere is the following reaction of N2O5:

%/h

> 20

-2 -1 0 1 2 3 4 5

10⁵ 1

2 3 4 5 6 7 8

10⁵

SO2 conversion rate 2019, EMEP,4.10 min, median, max = 2.6, 7.0, 108.8 %/h

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

%/h

> 20

-2 -1 0 1 2 3 4 5

10⁵ 1

2 3 4 5 6 7 8

10⁵

SO2 conversion rate 2030, EMEP,4.10 min, median, max = 2.2, 5.6, 102.7 %/h

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

104

N2O5 + H2O 2HNO3. (7.6)

HNO3 formed in the reaction above is initially assumed to evaporate and will take part in the forma-tion of ammonium nitrate or coarse nitrate. Coarse nitrate is formed from reacforma-tions with sea salt and dust particles with HNO3 (aq = aqueous phase, g = gas, s = surface):

HNO3(aq) + NaCl(aq, s)  NaNO3(aq, s) + HCl(g). (7.7)

HNO3 can also react with aqueous carbonates such as dissolved CaCO3 and MgCO3 on soil particles to form coarse mode nitrate:

2 HNO3(g) + CaCO3(s)  Ca(NO3)2(s) + H2O + CO2(g). (7.8) The MARS module in EMEP takes care of the partitioning between nitric acid (HNO3) and (fine) nitrate in the form of ammonium nitrate:

NH3 + HNO3 ↔ NH4NO3 . (7.9)

Figure 7.3. Chemical conversion rates for NOx _NO3^total _{= HNO3 + NO3}^coarse _{+ NO3}^fine (range [1.5 6] %/h) from EMEP for 2019 (left panel) and 2030 (right panel). Conversion rates for 2030 are computed with projected emissions and an 'average meteo year'.

NO2/NOx ratio

Because the NOx species have rather different dry and wet deposition properties, the deposition properties of NOx are computed as a weighed average using NO2/NOx ratios and a (fixed) HNO2/NOx ratio. These NO2/NOx ratios are derived from observations as a function of atmospheric stability and trajectory length according to the classification of meteorological situations used in the model.

The spatially variable NO2/NOx ratio is defined in terms of a stability dependent factor and a spatially variable one:

space n n

eff

n x s r x s f

r, ( , )= ( , ) _ . (7.10)

Table 7.2 presents the data for r_n( sx, )and fn(x,s) for the different classes for both summer and winter seasons on the basis of five years of measurements.

%/h

-2 -1 0 1 2 3 4 5

10⁵ 1

2 3 4 5 6 7 8

10⁵

NOx conversion rate 2019, EMEP,4.10 min, median, max = 1.5, 3.9, 5.1 %/h

1.5 2 2.5 3 3.5 4 4.5 5 5.5 6

%/h

-2 -1 0 1 2 3 4 5

10⁵ 1

2 3 4 5 6 7 8

10⁵

NOx conversion rate 2030, EMEP,4.10 min, median, max = 1.5, 3.9, 5.0 %/h

1.5 2 2.5 3 3.5 4 4.5 5 5.5 6

105

Table 7.2 Statistical data on night time NO2/NOx ratios and relative occurrences of night time

hours for the meteorological classes used in the OPS model. The data are derived from LML observations at rural stations over the 1980-1985 period.

Period Trajectory

length (km) Meteorological classes

U1 U2 N1 N2 S1 S2

NO2/NOx ratio: rn( sx, ) Summer 10 0.78 0.78 0.78 0.78 0.78 0.78

100 0.78 0.78 0.78 0.78 0.78 0.78

300 0.78 0.78 0.78 0.78 0.78 0.78

1000 0.78 0.78 0.78 0.78 0.78 0.78

Winter 10 0.47 0.47 0.62 0.69 0.39 0.58

100 0.47 0.47 0.62 0.69 0.39 0.58

300 0.47 0.47 0.62 0.69 0.39 0.58

1000 0.47 0.47 0.62 0.69 0.39 0.58

Relative frequency of

Summer 10 0 0 0.61 0.61 1.00 0.98

nighttime hours: fn(x,s) 100 0.17 0.17 0.68 0.68 0.63 0.83

300 0.43 0.43 0.44 0.44 0.42 0.44

1000 0.43 0.43 0.44 0.44 0.42 0.44

Winter 10 0 0 0.66 0.66 1.00 0.99

100 0.25 0.25 0.71 0.71 0.77 0.92

300 0.62 0.64 0.74 0.63 0.64 0.63

1000 0.62 0.74 0.74 0.63 0.64 0.63 It can be concluded from Table 7.2 that it is more important to include seasonal variations in the parameterisations than variations in stability and/or mixing height.

The parameter r_n provides diurnal and seasonal variations in NO2/NOx ratios to some extent. In the OPS model, also a spatial variation is introduced. This spatial variation is derived from a map of annual mean (background) NO2 concentrations in combination with an empirical relation between NO2 and NOx concentrations (see Eq. 7.17). The spatial variation factor, fn_space, is computed as:

[ ]

[ ] [ ]

[ ]



 



 +

=

=

6 . 8

4 . 12 exp NO

65 . 0

NO NO

65 . 0

NO

2 2 x

_ 2

bg bg bg

space bg

fn ^, (7.11)

with [NO2]bg in ppb. The value 0.65 represents the average NO2/NOx ratio for the Netherlands, so fn_space

has unity value when averaged over the Netherlands. Equation (7.11) is applicable to annual mean [NO2]bg values greater than 10 ppb, a value exceeded for almost all areas in the Netherlands; fn_space has a range of 0.50 (urban areas) up to 1.2 (coastal area of Friesland).

106

Figure 7.4: fn_space as a function of background NO2 concentration.

Note that in the future, NO2/NOx ratios from EMEP will be used.

HNO3/NO3total ratio for deposition parameters

Because of the very different dry deposition properties of HNO3 and NO3-, the ratio fHNO3 between (gaseous) HNO3 and the total secondary compound, NO3total (= HNO3 + NO3-) is used to average dry canopy resistances of HNO3 and NO3-. This fraction is modelled as a function of the NH3 concentration according to:

[ ]

[ ] [ ]

( ) ^{[ ]}

44 . 0 3 3

3

3 3

1000

024 NH . NO 0

HNO

HNO

⁻

−





 



= 

= +

^bg

fHNO (7.12)

in which [NH3]bg is the local (prescribed) background concentration of NH3 in ppb (see section 7.3 for background concentrations).

Figure 7.5: fHNO3 as function of NH3 concentration.

The formulation of fHNO3 is determined by a best fit to NH3 and HNO3 concentration results of a 1D chemistry model applied for the typical Dutch pollution climate for a period of several months. Because of the relatively high NH3 concentrations in the Netherlands, we can expect higher nitrate aerosol

concentrations than elsewhere in Europe. This is what actually is seen in the EMEP network (Hjellbrekke, 1999).

HNO3/NO3total, NO3coarse/NO3total , NO3fine/NO3total ratios for concentrations

The secondary species that is transported in OPS is NO3total = HNO3 + NO3aerosol, where NO3aerosol = NO3coarse + NO3fine, with the fine fraction NO3fine inside PM2.5 and the coarse fraction NO3coarse in coarse PM (i.e. PM10 - PM2.5). A map of yearly averaged fractions HNO3/NO3total, NO3coarse/NO3total and

0 5 10 15 20 25 30

0 0.2 0.4 0.6 0.8 1 1.2 1.4

[NO₂] _bg (ppb) fn_space

0 5 10 15 20 25 30 35 40

0 0.2 0.4 0.6 0.8 1 1.2 1.4

[NH₃] _bg (ppb) fHNO3

107

NO3fine/NO3total from EMEP have been made available from the year 2014 onward and are used in OPS to produce output of concentrations of HNO3, NO3coarse and NO3fine separately. In Figure 7.6, we show examples for the years 2019 and 2030.

Note that in the future, the EMEP-fractions will also be used for the averaging of deposition parameters.

Figure 7.6. Yearly averaged fractions of sub-species of NO3total from EMEP for 2019 (left panels) and 2030 (right panels). Top row: HNO3/NO3total;(range [0 0.4]), middle row: NO3coarse/NO3total (range [0 0.4]), bottom

--2 -1 0 1 2 3 4 5

10 5 1

2 3 4 5 6 7 8

10 5 fraction HNO ₃/NO₃^total 2019, EMEP,4.10

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

--2 -1 0 1 2 3 4 5

10 5 1

2 3 4 5 6 7 8

10 5 fraction HNO ₃/NO₃^total 2030, EMEP,4.10

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

--2 -1 0 1 2 3 4 5

10 5 1

2 3 4 5 6 7 8

10 5 fraction NO ₃^coarse /NO₃^total 2019, EMEP,4.10

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

--2 -1 0 1 2 3 4 5

10 5 1

2 3 4 5 6 7 8

10 5 fraction NO ₃^coarse /NO₃^total 2030, EMEP,4.10

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

--2 -1 0 1 2 3 4 5

10 5 1

2 3 4 5 6 7 8

10 5 fraction NO ₃^fine/NO₃^total 2019, EMEP,4.10

0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1

--2 -1 0 1 2 3 4 5

10 5 1

2 3 4 5 6 7 8

10 5 fraction NO ₃^fine/NO₃^total 2030, EMEP,4.10

0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1

108

row: NO3coarse/NO3total (range [0.3 1]). 2030 are computed with projected emissions and an 'average meteo year'.

In document The OPS-model (pagina 103-108)