LRTAP-6-75.pdf (1.520Mb)

A SIMPLE LAGRANGIAN DISPERSION MODEL APPLIED TO SULPHUR POLLUTION OVER EUROPE

ANTON ELIASSEN AND JØRGEN SALTBONES

KJELLER, 25TH APRIL 1975

NORWEGIAN INSTITUTE FOR AIR RESEARCH

P, 0, BOX 115, 2007 KJELLER

NORWAY

CONTENTS

Page ABSTRACT ... 0 ••••••• 0 •••••••• C • • • • • • • • • • • • • • • 5 INTRODUCTION . . . • . . . 6 The da ta . . . . . . . . . . . . . 7

Description of the model . . • . . . • . . . • . 8 An example: Computed large-scale S02

and S04 plumes verified by concen-

tration measurements from aircraft ...• 11 Model estimates compared to observed

S02 and S04 air concentrations at

LRTAP sampling si tes . . . • 12 Dry deposition of S02 in Europe 1974

as estimated from computed mean

concentrations 13

ACKNOWLEDGEMENTS . . . . 15 REFERENCES . .. . . . . . . . . . . . . . . 1 7

Table 1 Figures 1-7

18 19 - 25

A SIMPLE LAGRANGIAN DISPERSION MODEL APPLIED TO SULPHUR POLLU'TION OVER EUROPE

Anton Eliassen og Jørgen Saltbones Norwegian Institute for Air Research

P.O. Box 115, 2007 Kjeller Norway

ABSTRACT

A simple Lagrangian dispersion model is described and applied to sulphur pollution over Europe. The model cal- culations are based on available S0₂-emission data for Europe, and wind observations in the 850 mb surface. A special case is reported where the presence of computed large-scale S02 and S04 plumes are verified by concen- tration data from aircraft sampling and from the OECD sampling network. For stations in this network, computed and observed daily mean S0₂ and 80₄ concentrations are

compared for a period of six months. Based on this compari- son an S02 dry deposition pattern for Europe for 1974 is calculated, using computed mean concentrations and a depo- sition velocity of 0.8 cms-¹•

INTRODUCTION

As part of the OECD-project "Long_Range Transport of Air Pollutants'' (LRTAP) a network of sampling stations have been set up in the participating countries. The data ob- tained from this .network are daily mean concentrations of chemical components in precipitation and air. Two components in air are measured: SO2 and particulate SO4• Surveys of anthropogenic SO2-emission within Europe have been carried out in connection with the project. Atmospheric dispersion models have been developed to link the emission surveys and

the observed concentrations.

In the following a simple Lagrangian-type dispersion model is described. The model includes a transformation SO2 ⁺ SO₄ and thus gives estimates of SO2 and SO₄ air concentrations.

As an example the model calculations are compared with ob- served concentrations in a situation with large-scale SO2 and SO4 plumes crossing the North Seu. In this case the

data from the LRTAP sampling network are completed with con- centrations measured from the NILU aircraft.

In addition, calculated concentrations are compared with

observations from the LRTAP network for a period of six months starting from December 15, 1973. Based on this comparison a SO2 dry deposition pattern for Europe 1974 is calculated, using computed mean concentrations and a deposition velocity of 0.8 ems -1 .

The data

The SO2-emission data used in this work are based on avail- able information from the United Nations Economic Commission for Europe, and from OECD's Air Management Sector Group. A grid map giving the figures have been published elsewhere

(Eliassen and Saltbones 1975). Better information has now been received for most of the countries, but a complete survey was not available for this investigation. ·The yearly emission data are believed to be within± 20% from the actual figures, but may be somewhat more cincertain for the Eastern European countries. No seasonal variation has been included in the emission figures.·

The air concentration measurements within the LRTAP network are carried out by laboratories in the participating countries, using sampling and analysis methods specified for the project.

The detection limits have been estimated to 2-5 µgm-³ for SO2, and better than 1 µgm-³ for particulate SO_4• Locations vf the sites used in this investigation are shown in Fig 1. The geo- graphical coordinates of the sites used in the six months comparison with model calculations, are given in Table 2.

The wind fields used for advection are based on wind obser- vations in the 850 mb surface at 00, 06, 12 and 18 GMT. To obtain gridpoint values, the two wind components are analysed independently. The time interpolation between observation hours is linear in each component.

Description of the model

Consider a layer of air of thickness h flowing horizontally over a flat surface. Assume that the air has a constant den- sity, that the wind does not change with height, and that the S02 and S04 are completely mixed up to the height h. The

equations of continuity for S02 and S04 within the layer are

~=

dt

E - F

q q (l)

Ds

dt

=

^{Es - Fs ( 2)}

where q ands are the concentrations of S02 and S0_4, and E, q are source and sink terms for S02 and S011• The E F

s' q' operator

F s

D

dt denotes the total time derivative along a tra- jectory. The quantities in equations (1) and (2) are inde- pendent of the vertical coordinate.

The S02-emission term E is put equal to Q/h, where Q is the q S02-emission per unit area and time at the current position of the trajectory, taken from the emission inventory referred to earlier. For this work, the emission map used earlier, has been transformed to another grid and extended somewhit towards the east. Both grids have a grid distance of 127 km at 60°N.

No attempt is made to describe in detail the various trans- formation and removal processes of S02. The transformation

SO2 ➔ SO4 is assumed to be of first order, and the removal rates of SO2 and SO4 are assumed to be proportional to the concentrations. With these assumptions the equations (1) and (2) become

~

=

^{Q - kq (3)}

dt h Ds 3

ktq K·s _{( 4)}

=

dt 2

kt is the transformation rate for SO2 ^➔ SO_4, and k, Kare removal rates for SO2 and SO4• The factor 3/2 is the ratio of molecular weights of SO4 and SO2. The following values were used for the constants:

k

=

^{10-5 s-1}

k ·= K

=

^{10-6 s-1}

t

h

=

¹⁰³m

The authors have earlier (Eliassen and Saltbanes, 1975) re- ported some estimates of kand kt using

a

method based on

trajectories arriving at LRTAP sampling sites. These estimates were on the average about twice as large as the values given above. When complete vertical mixing of SO2 up to the height his assumed, a deposition velocity vs= 1 cms-¹ gives a removal rate v /h

=

^10-5s.

s

In the model, isobaric trajectories for marked particles are computed using the observed and analysed 850 mb winds. The S02 and _S04 concentrations q ands associated with the marked particles change according to equations (3) and (4). At the start of the integration, the number of marked particles is equal to the number of emission squares (32x32), and each marked particle is positioned iri the ~iddle of an emission square. New positions for the particles are calculated every

~t

=

1 hr, using a method described by Petterssen (1956).

Every 12 hours, 00 and 12 GMT, the integration is restarted with new marked particles in the middle of the emission squares.

By this time, about 15% of the old particles have disappeared across the grid boundary. The S02 and S04 concentrations of the new particles are obtained from those of the remaining old particles by an interpolation procedure, treating the directions paralell and perpendicular to the trajectories differently.

Model estimates of daily mean S02 and _S04 concentrations at a sampling site are obtained by averaging the estimated con- centrations of the timesteps covering one day. The concen- tration estimate at a certain timestep is the mean value of the concentrations associated with the particles present in- side a circle around the sampling site with the same area as an emission square. If no particles are present inside the circle, the concentration estimate of the previous timestep is used.

An example: Computed large-scale SO₂ and SO,, pl urnes verifj ed

!?Y

concentration measurements from aircrnft

As part of the LRTAP programme a number of concentration mca- suremen~s from aircraft have been carried out. When comparing these measurements to model estimates, one should expect the best correspondence where using measurements taken over the sea, where the anthropogenic SO₂-emissions are negligible. The , vertical concentration distributions may therefore closer ap-

proach the completely mixed conditions assumed in the model.

This also makes the wind in the 850 mb surface a more represen- tative advection wind for the layer. An example is shown on Figures 2 and 3.

The figures show the computed SO2 and SO4 concentration fields at 12 GMT May 10, 1974. A low pressure cell approaching from the west has set up a southeasterly airflow across the North Sea. The concentration measurements made with the NILU aircraft ar~ shown on the figures, together with daily mean concen-

trations from the ground sampling si tes of the LR'I'AP programme.·

The flight height was around 550 m.

It is seen that the model in this case gives about the right concentration levels. The observations confirm the existence of the computed SO₂ and SO4 plumes 500 km away from the closest upwind anthropogenic sulphur emlssions. Possibly a slight dis- placement of the cornpu_ted plumes towards the left would fit

the aircraft measurements better. This is consistent with barotropic boundary layer theory since the sampling height is well below the 850 mb surface, whe r e the winds used for advec-

tion are observed.

Model estimates compared to observed SO2 and SO4 air con- centrations at LRTAP samplinq sites

Model calculations have been carried out covering a period of more than one year, starting from December 15, 1973. The model

estimates are compared with observed concentrations from the first six months of this period.

In Table 1, the computed and observed six-monthly mean values of SO2 and SO4 air concentrations at 29 LRTAP sampling sites are listed. The table also gives the correlation coefficients between observed and computed daily concentrations in the period. For most sampling sites the number of daily concen- tration pairs were between 180 and 170, except for D2, D3, DK4 where the numbers were around 160, and DI<6, NL4 where they were around 150. The SO₄ correlation coefficients range from 0.241 to 0.775. The corresponding coefficients for SO₂ range from -0.019 to 0.610. At all sampling sites except two, the SO₄

correlation coefficienets are higher than the SO₂ coefficients, even though the transformation SO₂ + SO₄ is described simply

as a first order reaction in the model. Some explanation for this may be provided by the frequency distributions of ob- served and computed daily concentrations. At the site UKl

for example, (fig 4) S02-concentrutions lower than 16 µgm-³ are much more often obse~ved than computed. The model, in which complete mixing in a grid volume is assumed, is unable

to explain the observed low S0₂-concentrations in areas with large emissions. In these areas, the S0₂ is far from being uniformly distributed within a grid volume, because a signi- ficant part of it is emitted from point sources, as seen from a horizontal scale of 127 km and a vertical scale of 1 km.

For the S04, the mean transformation rate is slow enough to allow time for a more thorough mixing. Therefore, S0₄ is more uniformly distributed in the atmosphere than S0_2, and behaves more according to the model assumptions.

Factors like precipitation, vertical concentration gradients and wind shear are not included in this simple advection model.

This limits the day-to-day agreement obtainable between ob- servations and model estimates.

Dry deposition of S0₂ in Europe 1974 as estimated from

~omputed mean concentrations

Figures 5 and 6 show the computed six-monthly mean concentra- tions plotted against the observed ones (data in Table 1). De- noting the observed and computed S02 six-monthly mean concen- trations by y and x respectively, the linear regression line of yon xis:

y

=

^0.603x + 1.85 µgm-³ ( 5)

with a correlation coefficient of 0.935. Assuming random samp- ling from normal populations, the 99% confidence limits for the regression coefficinet are 0.603 ± 0.125.

The correspondence between computed and observed SO4 six- monthly mean values is not as good ·(Fig 6), even though the day-to-day correlation is better than for SO_2• Evidently,

the low mean values are overestimated and high ones underesti- mated. A larger value of Kin equation (4) would better

this situation, as this would reduce the low computed values relatively more than the high ones. The overall SO₄ concen- tration level can be adjusted by means of the transformation rate kt.

The good correspondence between computed and observed six- monthly mean SO₂ concentrations encourages a calculation of

an SO₂ dry deposition pattern in Europe for 1974. The yearly mean concentrations of SO₂ for each emission square is cal- culated from the model concentrations at 00 and 12 GMT each day. To transform these to ground level concentrations, the computed values are adjusted by means of the line y

=

^{0.719 x,}

instead of using the linear regression line (5). Both lines are shown on Fig 5. To obtain the dry deposition flux, a deposition velocity of 0.8 cms-¹ is employed, a value estimated by Owers and Powell (1974) to be representative for the British Isles

(referred to concentrations measured 20 cm over the surface).

The resulting deposition map is shown on Fig 7. The map differs somewhat from the corresponding ones calculated by Bolin and

large emissions and higher ones far away. The calculations of Bolin and Persson are based on a statistical formulation of the dispersion equation, and are valid for an arbitrary ye a r ,

ACKNOWLEDGEMENTS

This work was carried out in connection with OECD's Coopera- tive Technical Programme to measure the Long Range •rransport of Air Pollutants (LR'rAP) .

The following laboratories have carried out the air concen- tration measurements and have kindly given us permission to use the data.

Institute National de Recherche Chimique Appliqu§e, De- partement Pollution des Atmospheres (Dr. M. Benarie), Vert-le-Petit.

Umweltbundesarnt; und

Deutsche Forschungsgemeinschaft, Kornmission Luftverun- reiningende Stoffe (Pilotstation Schauinsland/Schallstadt

(Dr. G. Ronicke).

Eidgenossische Materialprlifungs- und Versuchungsanstalt fur Industrie, Bauwesen und Gewerbe, Zlirich.

Rijks Instituut voor de Volksgezondheid, Bilthoven.

Department of Trade and Industry, Warren Spring Labora- tory, Stevenage.

Finnish Meteorological Institute, Helsinki.·

Swedish Water and Air Pollution Research Laboratory, GBteborg (Prof. C. Brosset).

Danish Meteorological Institute, KBbenhavn.

The authors would like to express their gratitude to

The OECD programm e and the Norwegian Institute for Air Research for permission to publish this paper

the above mentioned institutions for permission to use their data

to Jack Nordø for valuable advice

and to the Norwegian Meteorological Institute for access to meteorological data and computer facilities.

REFERENCES

Bolin, B. and Persson, C. 1974. Regional dispersion and de- position of atmospheric pollutants with particular appli- cation to sulfur pollution over Western Europe.

Re.paid. AC--?. 8, In,5:l-J.:u;te. ofJ A{e;te.oJtology, Uiu,vvu.,d:y

06

Stoc.klw£.m.

Eliassen, Anton and Saltbones, Jørgen 1975. Decay and trans- formation rates of S0₂ as estimated from emission data, tra- jectories and measured air concentrations.

Atmo1.:iphv1,{,e, Envbwnme.n;t (.to appe.CUI..).

Owers, M.J. and Powell, A.W. 1974. Deposition velocity of sulphur dioxide on land and water using a ³⁵S tracer method.

Atmo1.:ipheJ1,{,e, Envbwrnne.nt 8, 63.

Petterssen, S. 1956.

WeAfhVL ana£.y1.:,,u., and fJ0Jte.c.a1.:i.:U,ng, Mc.GfLc<UJ-/·/iU, p 27.

Ul ro (I)

(I) .p

.j..) .. Ul

·r-l ro ^{u, -r-]}

Ul (I) rl r-l

::>

b1 ~~ H <!J S::: ^{(I) (I)} H

·.--l Ul .Q ro

r--l ,Q 8

Pa O (I) Ul

E: u ^(I)

rJ reJ (I) .µ

Ul S::: Q ·.--l CU Ul

~ ty,

,:r: -o s::: ^ty, E-i (I) ·.--l s:::

p:: .j..) .j..) ·.--l ....:l;jl-Jr-l

P, cu 0...

.j..) f. ^~J E:

rclOUlrcl

U Ul

l/J

s:::

0

·.--l .j..)

rv H .p s::: ^(!J

Q.l Ul §

u ^{.µ (l)}

S::: S::: Ul 0 (I)

0 ·.--l (I) ·.-l t) ,.C: ro ^N

S::: ·.--l ~J ~ I 0

få\µ QC/)

(I) lH S::: 0

E: ^(I) ·.--l O Ul

0 (tj

Ul O Ul 0

,.c: s::: ·.--l"' .P S::: 0 ,.C: I

s::: ^{0 ·.--l 0...} E:

0 ·.--l ~u ru tJi

E: -P ru :, .. ^1;1

I ro H ty, Xr-l-POS:::

·.--l (I) s::: ^{(I) ·.--l}

U)I-J(l)tyl

H O

0 S::: (I) H 0 0 ,.C: m

u ^~_J

s::: ro ^(I)

(l) 0 ..c:

(I) ·.--l .j..)

~ H

.j..) (I) 4-l (I) 0... 0

,.Q .-;-

Ul 0

OJ (.I)

.µ

(tj ~I

s::: 0

ro s:::

ro

(I)

Ul

s:::

000

(.I) ·.--l ·.--l

s:::

(I) N

ru:>os:::

(l) •.--l (.I) H .j..) ty,

;j

0-,Qri('")

8 Ul ·.--l r-- r-l Ori rumrl U ,:r: 'U r·l r-r:

.

~ E

~ ,

4-1

0 (1) M 0 0

"' ^{M ,,., co} '° " ^{0 ,n M ,n 0 °' r-- r-- 0 o, L') 0 ,n lfl} ""' ^{0 "1' co lfl} '°

-o r-- 0 ..,. ...,. ,-j

'° °' ,...

'° ^{M N M ,-j N N M 0 ,-j 0 N lfl r-- N M}

If) ::l N r-- .-l N N Ln ,-j q ,-j ,-j ,-j ,-j ,-j N

(1) .µ ,-j ,-j

.µ ·rl rj If) .µ

r. CJ .-l

·ri -1J ,:;

'tJ ·rl

4 01

0 (1)

0 _{0, '} 'tJ w w ::: w tel w w w ld t,1 ul w w w w w w tel w w w w w w tel w w :;:: :;::

u r: ::l

·ri .µ Ln r-- co

'° . '° ^r-- "" ^{L') N L{) L{)} °' '° ⁰⁰ '° ^{co 0 r--} °' ^{co ,n r--} °' ⁰ ..,. M

°' '° ^N

U--' ·rl ^{,q •} L{) ti) (') rl .-l

""' ^{0 N .-l M ,n <'l 0 ,-j M} "<l" "<l" N ,-j 0 ,-j L{) M 0 ,-j M 0 ,-j

·rl o. ^t;~

.c E: C 0 r--

'° ^co °' ^{.-l 0 Ln N co r-- L{) 0} °' ~, L'l

'° ^{SJ' ,n} "'1" r-- L{)

°' ^{M 00} "" '° ^{0 M}

P., rJ 0 .-l ~, .-l .-l .-l .-l ,-j .-l ,-j ,-j N N N N

rv If) .-1

).;

tr• 0 ^;J co co ..,. _{r-- .-l 0}

"" ^{0 N ,,..,} °' ^.-l ""' co r-- co C'> L{) co

'° ^{,o ,-j ,-j} °' ..,.

~·

^{N co °'}

(1) c;l -<!" ,n 0 0 N 0

"" ^{0 M N .-l} "" ^{0 M N L{) q L{) N} ..,.

"'" ^{lfl .-l "'1' M M N L{) , ...}

l'.) .-1

N r-- N ,...

'° '° "" ^{lfl co co ro co} °' ^{co N .-l N N .-l co} °' ^{M 0 0 , ... N} ,... ,-j Ln

z ^L{) "" '° ^{Ln Ln ,n L') L{)} "' ^{U) Ln L{) L{) Ln} '° ^{Ln L{) L{) Ln L{) L{)} '° '° '° '° '° '° ^{L{) Ln}

IC .µ

0 r.

·rl (lJ

;J ·d .-l <r

°' '° ^{co r--} "' ^{L{) ,n} ,... co N

°' °' °' ""' '° ⁰ .... .-l L{) L{)

°' ^{M 0 r-- 0 0} °'

c;l u ..,. L') L') Ln

°' "" ^{00 <r r--} "' ^{C•) M co ri} '° ^{·,o ri co} '° ⁰ °' ^{M r-- ID} °' '" ^{'<l ' 0 co}

ri ·rl N M M

'° "" ~,· "" ^{N r-- ,o L{) Ln "SI'} ..,. ID L{) ID L{)

'° ^L{) ""' ""' ""' ^{M M} "" ^M ,...

'°

(lJ ll-l

).; "" 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

,., (1)

0 0

u u

If)

C f'tJ 0 (lJ

0 ·rl .µ N L{) co

°' ^{N N co} ,- ^{N ._,.} "" ^{N N M} "<l" ID ,-j .-l •D ri co 0 r--

'° ^{l() (')} '° ^{M 0}

U) .µ ::l

rJ 0, r-- r-- N ..,. Ln L{) lfl Ln

'° ..,.

"' ^,q• "" "" ^(') '° '° '° '° ..,. M M M M M M

"' ^{,n L{)}

).; E:

.µ 0 C: u

(i

^{'tJ (1) >} ,., ^{N °' L{) M °' co 0 M r-- ,-j '° Ln 0 ~, r-- ,-j L{) °' 0 M L{) r-- N} "" °' °' "" ,... 0 C CJ ^L{) M 0

'° ^r-- '° ^co °' ,... L{) ..,. ..,. r-- Ln ,-j N co 00 ri Ln ,n N N N N ~, ,-j ,...

'°

nJ If) .-l ri ri

Q .Q

~ 0

C: .µ

0 C

·rl (1)

+i ^·rl "" ^{M .-l} '° '° ^.-l °' °' ^{L{) 0 ID} °' ..,.

"" ^\D r-- r-- ro "" ⁰ "" "" ^{N \D co co \D 0}

<tl u °' ^M ..,. u:, 0 N ,-j u m ,-j co

'" ^\D '° ..,. •r N

'° ^{"SI' M} °' ^co "" ^co °' ^{ri r, .-l}

.-l ·rl ri 0 1 ri 0 0 M 0

'° ^...,. M M N N Ln M L') M Ln ..,. ₀ ,-j N N <J' "" ^{M N} '°

(1) 4-1

,,

40 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

).; Q) 1 1

0 0

u u

If) C: -ø

"' ^{0 CJ}

0 ^{.,..; .µ} co

°' ^.-l ri lfl N

°' ^{N M r--} °' ..,.

°' ^{r-- N} °' °' '° ^{0 co N} °' °' '° ..,. co 0 'Sl' Ul .µ ::l 1

,"J C'~

°' ^{N 0} "" ^r-- °' '° '° ^co ,...

'° °' ^{co M co .-l} '° ^{co 0 r-- M} '° ^r-- '° ^{'Sl' ~1 r-- co}

).; F. ^{M N ,-j} .-l .-l .-l ri N M M N M ri M ri

.µ 0 C: u

CJ u r: 'tJ 0 ^C)

u > ,., ^co '° 1 ^L{) "" ^{M N M r-- ,-j} '° ^{co L{)} °' ^{M M r-- ,,.., L{) N .-l ,-j L{) N ,n co ~1 'Sl' r--}

c: (1) Ln

"" ^{r-- r-- ,-j "' 0} "" ^{co L{) Ln .-l r-- M "1' r--} "" ^{,-j \D L') M} '° ^{,n ø L{) L{) (') N}

i,j Ul N .-l .-l .-l N .-l N ri ,-j M .-l N .-l

GJ .n

"-• 0

.,.

N M ri N M <r ^L{) '° ^{,-j ,-j M} °' ^{N M ,n .-l N M} .,,. ^(') <I' L{) ,-j N M ..,. L{) ,-j N :,<. :-<: :r. ,c :-<: ::,: _{N N N} .-1 .-1 .-1 ,-1 '4 u, k, '4 k, :,,:; ::,:

0 0 0 0 0 0 ~ ^{c:, k,} 7. z 7- z 1/. ?, 1/. -, ~- _{7, :;c. Ul U) U) Vl U) U) Ul (/J ;:i ;:i}

~eu:~~~~4,_~~.b:~~~~~·~~¥C,::7~~-~.£'"'~'!»!.""!.'i..;,.;..7Ji;

If) (1) If)

nJ u

).;

Gl 4-1 æ

;J ::l .Q

ri .-1 z

40 0

Q) If)

.c: 0 .µ Q ).;

cu

If)

•'

Q) nJ E

·rl .µ Ul (1)

.c: Q)

E-!

,---j (I) r-l ,Q E-i ro

+

OTHER STATIONS

DK1 ~

a. ,:,__,,,'

I

0D , ~.,• ~ '

d I \

Ø l

I

I

. i, S.5 '

-

^.

I I

\

N25 8 I I

'

I I

u~1. +

~UJ{1~

F.5+

f1

f)

+

f6

f3

+ +

F2

•

•

'

^{.... ...}

..._

- -

^..._

' 'I

'

_\.

' _'

^{\ I}

I I I I /

9

SF5

9

' I I

\ I

'

^'- . '

I

I

\

' I l

-

^....

\

I I

SF3_o ¹ _I

I

\

\

F4

+

04

+

' I

I I I

I

I

'

'

I

'

^\

'

^{' ~}

,J,.- _ _,,,,,..--, /,,

I I ' ..,.

/ (

\ I

I /

I I

. I

I

\ I

Fig 1: Location of LRTAP sampling sites providing data for this investigation. Data from stations marked with+ appear only on Figs 2 and 3.

' . ^'

J - • 2

' _\

.

¹¹

5

10

, 1,4

19v§

l •

~do,;-o '· ~'//, ~

JO •

·•JC/~-

) • 40 . ,'

so '

' L

' .,,,. <l. ø •

' ,, -

"'~

8.

~,.b \

~-

\

--

\ - ,

\ ,, ,,

"

\ I

·o

·o

.o

©- x--x FLIGHT-TRACK

Fig 2: Computed SO₂-concentrations at 12 GMT May 10, 1974, (isolines) together with aircraft measure- ments (in circles) and daily mean concentrations from the LRTAP-network. Unit: µg SO₂/m³•

• 0,0

®- ')( -FLIGHT-TRAC)\

Fig 3: Computed S04-conccntrations at 12 GMT May 10, 1974, (isolines) together with aircraft measure- ments (in circles) and daily mean concentrations

from the LRTAP-network. Unit: µg S04/m³•

'l

~--

^f

e'

0 0

w ^w

(J)

0 ^{>· !-}

~

LlJ ^a.:

::.) N LLl 0... r--

er:: ^{l/) ~}

LU ro ₀ 0

...J

1--

u <!

~- I I ^,-

⁰

C) ^{I- ~}

0

z I

,;;-

X ^N

:J

0

V)

(J)

N

rd

Q) •

.w r-l ::J ~ 0...::, F-: 0 Q) 0 .+J

·r-1 rd _{/)

i::

((i (I)

..c:

rd .+J

Q)

::,. .+J H ro

Q) UJ {I)

.Q i:::

0 0 ·r-1

~ .+J

o ro

i:: .j.J H 0 i::

·r-1 (I)

.+J 0 ::J i::

.Q 0

·r-1 0 H I .+J "1

{/) 0

·r-1 CJ)

rd i::

~ ro

0 (1)

i:: I::

Q)

::J ~ tJ1 ...-i (1) ·r-1

H ro

J:t.ird

tn l.J.J

~ ~

j.::

li)

Lt.I 0 z

l.l.J ^\J)

z ^7.: ₀

_ _J

I-

z ^<i: _>

0 er

I- w

u ^li)

[Jj rJ.l

CC 0

o: ..

0 z

u ⁰

C) (/)

ltJ ^Ul _LI.J

....J er

0.. L) 0.. w

~ (.L

I

I

^{~ N}

LU 0

~3

{)_

'<C' ..,f: ~-

0 u

:..'✓ I

--

_{:-;_} E ^CJ!

N (V)

s:::

{/) {/) ..c QJ

i:: (i) ~

0 .µ

·ri H etl ro +JetlE;GJ etl QJ ·ri ·ri HC+lrl

~J ·ri {/) 0.1

c ^r-1 GJ 0..

QJ etl

O QJ c

i::..coc

0 8 0

0 Ul ·ri I • c .µ

N {/) Q 0

0 GJ ·ri a>

Cl)C+JH • 0 etl H ,::

c :> 0 0

etl ro H O ·r-1

Q)Q)Q) .µ

Ei :> ^Ul ,.C ·ri H.D.+l Ul :>-t QJ O ·ri 0 ri Ul ;3 0..

,.C ..Q lH QJ

+lOOl--lro

c GJ

O+JQJ..C:>-t E':UlC+lH I ~ •ri CiJ ro

~ ·ri ri tJ) -r-i n:l O tJ)

Ul tJ),:: .µ S:::

rel O ·ri ro -r-i s::: .µ CiJro!./l~rtl +JQJUlOrl

;j .µ Q) ..c ;j 0..+J H Ul 0

801:J) ri

Orl<lJUlrt:J U P.., H -r-] O

11)

co ⁰

OBSERVED

so,

µg/_m-3

16

· 12 +

+ ⁷

+ +

8 +

Ji..;.

-1- + + + +

++ + +

++

4 +

++ _ _j + "1'

+

I

JJg/m-~ COMPUTED S04

4 8 12 16

Fig 6: Computed six-monthly mean S04-concentrations plotted against observed ones. The linear regression line of observations on estimates is shown together with correction line applied when calculating dry deposition.

6

I I

': ;·

\ 9-> /

I'-. r ~

-✓

..,... I

I

0

C>

.J ⁾

Fig 7: Calculated S0₂ dry deposition pattern for·Europe for the year 1974. Unit: g S02/m^2•