• No results found

Ödsmål, Kville sn, Bohuslän Hällristning

N/A
N/A
Protected

Academic year: 2021

Share "Ödsmål, Kville sn, Bohuslän Hällristning "

Copied!
12
0
0

Loading.... (view fulltext now)

Full text

(1)

Det här verket har digitaliserats vid Göteborgs universitetsbibliotek och är fritt att använda. Alla tryckta texter är OCR-tolkade till maskinläsbar text. Det betyder att du kan söka och kopiera texten från dokumentet. Vissa äldre dokument med dåligt tryck kan vara svåra att OCR-tolka korrekt vilket medför att den OCR-tolkade texten kan innehålla fel och därför bör man visuellt jämföra med verkets bilder för att avgöra vad som är riktigt.

Th is work has been digitized at Gothenburg University Library and is free to use. All printed texts have been OCR-processed and converted to machine readable text. Th is means that you can search and copy text from the document. Some early printed books are hard to OCR-process correctly and the text may contain errors, so one should always visually compare it with the ima- ges to determine what is correct.

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

CM

(2)

Ödsmål, Kville sn, Bohuslän Hällristning

Fiskare från bronsåldern

Rock carving Bronze age

fishermen

Hydrografiska avdelningen, Göteborg THE BRAl/D BLDUJ OUT. OIL POLLUTION

ANALYSIS ON NORTH SEA WATER 9-11 MAY 1977.

PRELIMINARY RESULTS.

Martin Ahnoff, Stig R. Carlberg, Lars Johnson, Bengt Yhlen.

J

June 1977

(3)

THE BRAVO BLOW OUT

OIL POLLUTION ANALYSIS ON NORTH SEA WATER 9-11 MAY 197?

PRELIMINARY RESULTS

t» y

Martin Ahnoff and Lars Johnson

Department of Analytical Chemistry University of Göteborg

Fack, 3-402 20 Göteborg

Stig R. Carlbarg and Bsngt Ybien National Board of Fisheries, Hydrographic Department

Fack, S-403 10 Böteberg

ABSTRACT

During a cruisa with R/V Argos in the North Saa 9-11 May 1977, water samples were collected at nina stations asst and north east of the Ekofisk area. Tha samples were extracted and ana­

lysed for petroleum hydrocarbons by spectrofluorimetry. At eight stations, no significant increase in the petroleum con­

tent could be noticed, compared with earlier measurements in the same area. At one station, about 60 nautical miles ENE of the Bravo platform, a tan times higher level was found, which extended from the sea surface down to at least 30 m depth.

No traces of oil were visible on the sea surface. No oil par­

ticles were found at any of the stations e.s judged from visual

inspection of net samples.

(4)

INTRODUCTION

Within the Ekofisk area in the North Sea a blow-out started at the oil production platform Bravo, situated at N 56*33' E 03 12.2 , at 22 April. The blow-out caused an uncontrolled outflow of oil during the nearest days at an estimated rate of 3 - 4000 tons per day.

At 30 April the out-flow was mastered and the oil well was capped. The oil was reported to have spread over an area of about 4000 km , although the oil had been broken up in several fields, intersected by water without visible traces of oil.

Because of the importance of the North Sea for, amongst other things, the commercial fisheries it was of course of primary importance to undertake.as quickly as possible a number of investigations in order to assess information about possible damages to the biological life in the sea area in question.

Several groups from the Institute of Marine Research in Bergen were quickly on the spot, and some preliminary results of their work have been reported (1).

The National Board of Fisheries in Sweden had a cruise for biological investigations in the Skagerrak and North Sea area scheduled for early May. The programme for this cruise was modified to include samplings for chemical analysis of oil in water and some net samplings of particulate oil as well.

Because of the limited ship time available, it was however, not possible to extend the investigations to the immediate vicinity of the Bravo platform.

Because the hydrographic department of National Board of Fisheries was in the process of moving to new premises, the chemical laboratory was inoperative. The colleagues at the Department of Analytical Chemistry at the University of Göte­

borg kindly put their expertise and analytical capacity at disposition and with joint efforts this report with prelimi­

nary results could be presented. The Institute of Oceanography at the University of Göteborg had a cruise at the western

border of Skagerrak at this time, and thus additional sampling

could be made and the results be incorporated here. In figure 1

the localities of the sampled stations are shown.

(5)

METHODS

Water sampling

The sampling and analytical procedures for water samples were indentical to those described by Ahnoff et al. (2) and Ahnoff and Johnson (3). Water was sampled at depths from 1 m to 30 m.

Closed 1-1 bottles were lowered to the desired depth in a special sampler (2). A teflon film covering the opening was punched by a spring loaded mechanism triggered by a falling messenger. The bottles were then filled with water in 5-15 seconds. Samples from 1 m depth ware taken by an open "drop bottle” of the type recommended for IGOSS (A). Both sampling methods ensured that samples were taken unaffected by the surface water or a surface oil slick.

Net-sampling and trawling

As a rough scanning for oil particles in the water, net-sampling and trawling were carried out on some stations (Table I).

A bongo-net with two openings of 600 mm in diameter and clothing with mesh size at 300 jim and 500 p.m respectively was towed in oblique hauls between the sea surface and 50 m depth. The towed distance was about 0.5 nautical mils and the filtered volume of sea water was about 250 m3„

An Isaac-Kid Mid Water Trawl (IKMWT) was towed in the same way for about 0.9 nautical mile, filtering approximately 20 000 m^

of sea water.

Analytical procedure

Water samples were extracted aboard the ship directly after collection, using n-hexane as extractant. (The solvent was

spectroscopic grade n-hsxane, Merck, which had been additionally purified on an activated silica gel column). The extraction was performed in the sampling bottles by means of vigorous magnetic stirring during 45 minutes, The extracts and possible emulsions were pipetted off and stored in test tubes and analysed upon ' return to the 1ab oratory on shore. Without further pretreat­

ment UV-fluorescense spectra of the extracts were recorded

with a double monochromator soectrofluorimeter. (Aminco Bowman

(6)

SPF) equipped with a 200 td xenon lamp. Ekofisk crude oil (more than one year old) was used as a reference substance.

For quantitation the following wavelengths were used:

excitation 230 nm, emission 340 nm.

RESULTS

Water samples

The measured content of petroleum hydrocarbons (see Tabe II) was very much the same at eight of the nine stations ( see Figure 1) the concentration at 1 m depth was 0.9 - 1.7 |ig/l and it decreased to 0.5 - 1.0 fig/1 at 30 m depth. In compari­

son, water samples taken in the North Sea in October 1975 at 2 m depth contained 1.0 - 2.0 ng/1 (see Table III). Compared in this way, no significant increase in oil content was obser­

ved.

fit one station (No 7) the concentration was 13 (xg/1 at 1 m and 6 jig/1 at 30 m depth. This is ten times the mean level at the other stations. It is also considerably higher than any con­

centration measured by our laboratory in open sea water using the same analytical method, including samples from the North Sea, Skagerrak, Kattegatt and the Baltic Sea.

Net samples

At visual inspection no oil particles could be detected in either of the samples obtained with the bongo-nets or with the Isaac-Kid trawl. The only exception was a trawling in the vici­

nity of station 1 which yielded some oil particles, which were hard and apparently of other origine than the Bravo blow out.

CONCLUSIONS

This investigation, although limited, shows that ten days after

the uncontrolled leak on the Bravo platform was stopped, at a

time when no visible traces of oil on the water surface remained

in the area of investigation, the polluting oil had spread into

the water body down to at least 30 m depth. At this occasion,

9-11 May, high levels were found only near the Ekofisk area

and had apparently not spread to the Norwegian coast or the

Skagerrak.

(7)

ACKNOWLEDGEMENTS

Thanks are due to 0. Hagström, National Board of Fisheries, for providing information about the results of net-sampling and trawling. The Institute of Oceanography at the University of Göteborg kindly put the results of their observations at our disposition. Thanks are also due to Dr. K. Palmork at the Institute of Marine Research in Bergen for providing the refe­

rence oil and to the officers and crew of R/V Argos for their extra work, caused by this additional part of the programme.

REFERENCES

Bratberg, E (editor) 1977: The Bravo blow-out. Fisken og Havet Serie B, Nr 5. Fiskeridirektoratets Havsforskningsinstitutt, Bergen (Mimeo).

Ahnoff, M. et.al. 1974: Report on the Chemistry of Sea Water XIV. Dept, of Analytical Chemistry, University of Göteborg, Nov. 1974 (Mimeo).

Ahnoff, M. and Johnson, L. 1977: Quantitation and characteri­

zation of petroleum hydrocarbons in Baltic Sea Water.

Report to the Soviet-Swedish Expert Meeting on Evaluation of Results from the Joint MUSSON-Expedition 1976 and Planning of Future Cooperation, Göteborg, April 24-30, 1977 (Mimeo).

Guide to operational procedures for the IGOSS pilot project on marine pollution (petroleum) monitoring.

IOC/WMO Manuals and Guides No 7, Unesco 1976.

0. Hagström personal communication

(8)

Figure 1

Stations worked by the Institute of Oceanography

1977.

NORGE

'** SHETLAND ISi.CS

Argos I975

ORKNEY JSlA*«DS

BRAVO Argos 1977

NORTH sea

D

anmark

BUNDESREPUBLIK DEUTSCHLAND ENGLAND

BELGIQUE

FRANCE

0*

8'

(9)

Table I

Samplings performed

Argos 1977 Water Bongo-net IKMWT Station no samples

1 2 3

4

5 6 7

a

9

x x x X

X

X

X

X

X

X

X

X

X

X

X

X

On all stations water was sampled at the fallowing depths 1 m, 5 m, 10 m and 30 m.

IKMWT = Isaac Kid Mid Water Trawl

*

(10)

CD QJ

•a -H

c a

ca E /-N ca ca CD eu eu

•H •p P oj QJ U JC •H CO H

•H a U. 3 4- TJ

o 4- TJ Q

P 03

HJ ai

a ta

m p

rH QJ >

CO CD c QJ

•H P

•P CD 2 CO CD w eu CD rH 3 a CD en >

h CE •—I

rH

S

Œ

CK .

sz en p h •H O

3 S3

en

p03 B :□

en ta

M «H

M 3 ».

rH en p

c •H

>H CD P P 3 QJ TJ >

C •H C eu 3 jy OJ p

en SZ a

m CO rH

p

a ai E a CD c CD QJ P QJ U QJ O P CD 4- 3 E OJ CD P

P 3

P P

TJ •H P QJ CD C C

*H M CO V XJ P U a •H CD C co p 3>

rH en 3 CD 3>

eu S CK CK

rH

\ CM O UJ o

ai

3 «C* a

JC p E E E E a QJ r en a a

Cs V m

£U -P

ra

tj TJ C

CO en r*

c » » 1

a c oj a LH

•H a □ CM a

P •H O 0 i

•H P <n un o CD CD un o o

D P erv

a. en 2 LU r*

rH

S C\J

en un en

ai

3. r- O

a

JC p E E E E eu a r- en O a

Ca

v-

m

P

B

CO

TJ

TJ

C

CD C en

s r o v- O c O -4 en 1

•H a en en a

P -H o o i

«H P en m C- CO CO en a o

a p en

CL en 2 UJ r*

rH

S O DJ en cs

CJ1 *

3 ^r*~ r- o

x: p

m m

E E eu a. <p~ en O O

o r* m

0!

pCD T!

TJ C

QJ % » en

r- o a

a

C • •

c

un m en

•H

a

O J-

o

P •H o o i

•H P

en

en D' CD CO

un o

C-

a

p en

CL

en

2 UJ r-

<r -3- m

• • •

O v- r-

E e E t- □ a

'T- DJ

\ K -4 a

a

a

m

i 3- CM

a

2 o o i LU co en

o

CK in a

c'­

XC en

n

2 en

r-

en O en un

«

o O o o

E E E E

'f- m o o

v n

UJ

\ \ en un

1 0

C

• ■

en un i a

in en

uno

•H o o . i P en in o

CD un o o p

en

2 LU en

r-

tn O o c-

* *

o a

E E E E

en 1 0 a n

CM

1 0

v » 1

c

co en

un a

n -4

o

•H 0 o i p D- CO

er-

CD un □

o

P

en

en

2 LU s—

N D D

E E E

t

- a a

r- DJ

S

_l

a

u * UJ X

un uj CM

• 1

cn o m

-4

2 m

cm

a ce

o o t X

O CO o

m a

C'­

en

2 2 Lü

r-

n r- en CM

• • •

n

V-

CO UJ UJ

m m E E

r- un a D r- n

r- a

>. s r~

c UJ -4 en 1 a -4* r* D

•H

0

o i p UJ j- D-

ca m en D-

p cr>

en 2 LU r-

m en un

« •

*

r-

r* CD o

E E E E

en O a r- m

m S V CO CD

1 0

c

• • CD en I a en n eno

•H o 0 i P r- c'­ e>

£0 en o r- P

en 2 LU cn V“

in -d-

• e

E E

v- n DJ

s:

a un

X

a e— i en en

2 a

cC i X r-

*

C'­ en 2

r-

a en CO a

»

r- a a

E E E E

t

- un a a

t- m

co

c

r rCM -4

5 -1 1

o

<c-

m o

•H o o

i

P O -4

o CO m a er- en p

2 LU

en

r-

CM

en

CM

o

• • •

r-

o a

E E E E

'f- un a a

V- m

* O

-4 CM v-

• « ! C en D- en a T- -4 a

•H

0

o i

P en co c—

QJ en o ir-

P cn

en 2 LU V“

(11)

Table III

Resulta obtained from water samples taken at 2 m depth during a cruise with

R/M

Argos 14 - 16 October 1975.

Position 49/1

N 59 °20' E 04 °50' 1.7

It

04 °3Q' 2.0

ff

04°11' 1.7

It

03 °51' 1.8

tl

02 °Q0' 1.7

II

01°19' 1.4

II

00 °40' 1.1

ff

00o00' 1.0

N 58 °30' W 02 °51' 1.3

(12)

References

Related documents

Figis 5 and 6 shows the transversal ana longitudinal distribution respectively in the Bothnian Bay in June 1972, Here we find the highest open sea values of yellow substance;

Hydrographical measurements performed by the Fishery Board of Sweden in investigating the Brofjorden and primary production measurements by The Marine Botanical Institute,

The water samples for total phosphorus were collected once a day at 07 o'clock GMT and sent once or twice a month to Danmarks Fiskeri- og Havunders/gelser, Charlottenlund, where

According to Swedish investigations the discharge of easily oxidable organic matter from communities cause first a primary oxygen reduction in the recipient when the matter is

4 show the oxygen and hydrogen sulfide conditions in the deep areas of the Baltic during 1976« Prom these it can be seen that there was hydrogen sulfide only in the eastern

In order to test the precision (or micropatchiness) and the depth distribution of phosphate in the surface water, a series of samples were taken at every meter down to 10 m using

surplus of Baltic water along the Sweaish Skagerrak coast, contemporous with a strong spring bloom of plankton. Also the autumn was characterized of low oxygen in the SE Kattegat

An inTQst.igat.ion of the oxygen and phosphate conditions in the Baltic (jfoaseliua 1968) shows a strong decrease of the oxygen concentration and a great increase of the