lCES C.M. 1986/C:7 Hydrography·Commlttee/5ess. Q ON LONG-TERM VARIATIlONS OF SOHF. HYDROGRAPHICAL PARAMETERS IN THE DEEP BASINS OF THE BALTIC PROPER BY ST1G H. FONSELIU5 Swedlsh Meteorologieal and Hydrologieal Institute Oeeanographieal laboratory, Göteborg P.O. Rox 2212 5-403 14 Göteborg, 5weden ABSTRACT In the Baltie Sea long-term series of salinity, temperature and dissolved oxygen' observations from the maln stations exist. The series begin at ·the end of the last eentury. Nutrients eoncentrations have been measured at these stations ~fter W.W.II. Phosphate measurements began in the 19503 and nitrogen compounds and silicate have been measured sinee the end of the 1960s. By help of these saries it has been possible to study long-term trends in the surface water and to follow the large variations in the deep water, caused by inflows of Kattegat water, with high density along the bottom of the Baltlc proper. Unfortunately no similar long-term serles of biologieal parameters exist. Therefore it 15 diffieult to draw reliable conelusions regarding biologie al trends and variations and to couple biologieal results to the hydrographieal changes. " • In the present pap~r series of density and oxygen measurements from the surfaee to the bottom at the station BY 15, the Gotland Deep are shown. Inflows of water with high density below the haloeline are shown to cause stagnation and hydrogen sulphide formation in the bottom water of the deep basins. The stagnation periods are broken by new inflows, improving agaln thc oxygen conditions. DIagrams for phosphate variationsfrom 1967 and nitrate variations from 1968 to present time in the Gotland Deep are also shown. The aeeumulation of phosphate and dlsappearancc of nitrate in thc stagnant water are demonstrated. Trends of some blo1ogical parameters are briefly discussed. The observed eutrophieat ion of the Baltic Sea surface water and posslble effects on reproduction of f1sh and also thc diffieultics encountcred in trylng to conneet fisheries statlsties to hydrography are finally diseussed. ON LONG-TERM VARIATIIONS OF SOME HYDROGRAPHICAL PARAMETERS IN THE DEEP BASINS OF THE BALTIC PROPER BY STIG H. FONSELIUS Swedlsh Meteorological and Hydrologieal Institute Oeeanographicnl laboratory, Göteborg P.O. Box 2212 S-~03 14 G~teborg, Sweden In the Ra1tie Sea we find long-term Baries of measurements of sa1inity, temperature and dlssolved oxygen beginning at the end of the last eentury. These series have been taken at the main deep stations rather regularly. The work has bcen carried out by marine laboratories from different countries Burrounding tho Baltic Sea. When the lCES was founded In 1902 thc work was by agreement divided between the Ealtic Sea countries in order to try to cover the main decp stations at least once every yea~. In the beginning the work could not be carried out very regularly. For some years data are completely miBsing. Often only one expedition was carried out annually. Unfortunately the werk hadto be Interrupted durlng the two World Wars (W.W.), due to military operations and lack of research ships. Therefore we find gaps of several years in thc ser1es du ring the war periods •. • After W.W. 11 the amount of observations has increased enormously and all the main stations are now covered at least four times annually. Afte~ the war, observations of nutrient concentrations have been introduced in all Baltic Sea countries. The analytical methods have been worked out. Intercalibrations have been carrled out In the frame of ICES und we now have quite rel1ablc ser1es of. phosphate measurements slnec 1954. In the middle of the 19605 nitrate and nitrite analyses were introdueed. Also silicate and ammonia have becn measured on a routine basis since the beginning of the 1970s (Fonselius 1986). Fig. 1 shows a map of the Baltic.Sea with the stations discussed In the paper marked. Fig~ 2 shows thc variations of density in the deep. water of the Gotland Deep (BY 15) from 1883 to 1985. The 'density 1s rna1nly dependent of the salinity in the deep water, where the temperntura variations are low and a d1agram of thc salinity would be very similar to .the density diagram. In thc figura we can see that an inflow of Kattegat water with high density occurred durlng W.W.I and that thc denslty then decreased. The high peak in 1928 1s caused by a sIngle set of measurements and 1s obvlously wrong. The measurements were carried out by the "Skagerak" and the next day thc same station was occupied by the ,.Aranda" , which got quite normal results. The "Skagerak" values have to be reduced with 0.5 salinity units from thc surfaee to thc bottom. My raason for this correctlon i8, that the oxygen values of the decp water were extrernely low. An inflow of high saline water would have contalned oxygen. Fig. 3 shows the oxygen variations in thc Gotland Deep in thc same manner. We can see periods wlth hydr~gen sulphide formation (the black arens 1n the fig.). These 3 are always corresponding to periods with decreastng density. Tho oxygen peaks correspond to inflows of water with high density. If we return to fig. 1 we can see that thera was another large water inflow In 1952. Thls Inflow was the largest ever obscrvad in the Baltic Sen. It has been descrlbed in detail by Wyrtki (1954). Both the above monttoned inflows caused hydrogen sulphide formation in the bottom water. After the 1952 inflow, increasing periods with stagnation and hydrogen sulphide formation have been obscrved (fig. 3). The inflow obviously was a catastrophe for the Baltic Sea. It has often been clairned that eutrophicatlon 1s responsible for the hydrogen sulphide formation in the Baltic Sea (Larsson et al. 1985). The stagnation pcriodn, howcver, are not caused by eutrophicat1on and the increasing occasions with salt water inflows followed by stagnation, is the prirnary renson for the bad oxygen conditions and hydrogen sulphide fo~ation. Eutrophication r.~y have increased the oxygen utilization, eausing more hydrogen sulphlde formation. . , • During stagnant, conditlons phosphate is accumulated in the stagnant deep water and when hydrogen sulphide 1s formed, phosphate is released from the sediments (Fonsellus 1969). Nitrate is also accumulated, as long as oxygen is pro3ent in the' water. ~oen the dissolved oxygen reach values below 0.5 mlll nitrate begins to be reduced to nitrogen gas and disappears completely from the stagnant ~~ter during reducing conditions, when hydrogen sulphide 1s ro~ed. Fig. 4 shows tho'phosphate variations in the Gotland Deep in the whole water muss from 1967 to 1985. We can see in the figure the accumu1ation of phosphate and the seasonal variations in the surface layer. Fig. 5 shows in the same manner the nitrate variations from 1968. We can see that the nitrate disappears in the deep water cornpletely or almost cornpletely. This 1s due to the analytieal technique, which 13 not very aecurate.Often tho zero results are reported as <0.1 and sometimes valucs are given with two decimals even below 0.10 unol/l, whlch i~ meanlngless. In the surfa'ce layer the nitrate disappears eompletely during the sumrr:er period, indicating that nitrate 1s one of the production limlting factors. • From these figurcs 1t is difficult to see if the nutrlcnt levels in the surface water have incrcased since the 19605. Nehring et al (t985) has shown that,the nutrient concentrations in the whole surface layer of the Baltie'Sea has increased during that period and that the salinity of the ~~ter also has increased. Figs. 6, 7 and 8 show examples of this increase In,the Gotland Deep, as annual mean values at 40 'm depth for salinity, phosphate and nitrate during the period 1960-1985. Unfortunately we do not have corresponding long-term observations of biologlcal parameters. We do not know if the primary production of the water has inereased, we also do not know if the blomass has increased. There are some signs of an increase of the zooplankton biomass, but also contradictory rcsults have been 'obtained (Wulff et al 1986). Schulz (1985) has shown that the chlorophyll a has increased in the southern Baltie Sea from 1975 to 1983. He eoüld not find satistieally significant Increases for pri~ary production and zooplankton biomass during the same period. One or the difrtculties in blo1og1cal long-term investigatlons is, that gear and methods have been changed several times and that the results therefore are not comparable. 4 For fisherics long serles of eateh statlsties cxist, but the 'results are hardly useful for conelusions rega:rding ehanges in thc flah stock. If we look at thc eonditions in thc Paltie Sea, eod was not caught in larger arnounts before W.W.II. It was not considered eommercially interesting. During W.W.II thc German trawling fleet was due to the war eonditions exeluded fro~ its normal fishing grounds In the North Sen. Therefore it moved into the Ra.ltie Sea. This gives a high response in eateh statistles, but of course the result depends on new fishing methods, trawllng was not usual in the Baltie Sea at that time, and of course ~n inercased effort (Mayer and Kalle 1950), (Otterlind 1986). Otterlind also points out other faetors, which heve inereased thc eatehes after W.W. 11 as inereased fishing efforts mainly by Poland and USSR uslng new large trawllng fleets, transfer of fishing efforts from the North Sea to the Baltie, ehanged market eonditions ete. (fig. 9). • • It fs not easy to conncet variations in flah enteh to hydrographie eonditions. We have to remember that there are saveral links in the food ehain between thc phytoplankton, whieh first 1s influeneed by ehanges in nutricnt eonditions i~ the surfaee water and eommercial fish, whlch 13 ineluded in flsherles statistics. Hydrographie ehnnges do not only have etfcets on the nutrient eondltlons. stagnation of the deep water may lead co anoxia and .hydrogen sulphide formation in the Baltie baslns. Thls may influenee the cod reproduet10n 1n thc Baltie Sea. Important spawning areaa are the Bornholn Basin and thc Gdansk Rasin, where anoxia ~ay develop. Svansson (1985) has tr1ed to eonneet good year elasses of eod with thc hydrographie conditions in the Bornholm Basin (fig. 10). From the flg. we ean see that very good elasses developed in 196 1., 1972, 1976' and possibly 1977. Good elasses dcvcloped 1n 1963, 1966, 1979 and 1980. Otterlind (1986) additional1y points out 1977 and possibly 1981 as years for good elasses. Very bad was the elass of 1968. The very ~cod elasses eoincide with inflows cf new ~~ter, whieh improved the oxygen oondltlons in the whole southern Baltle Sea, thus Improvlng the spawnin;g rcsult. One rnay or course point out that eod eaught in an area does not neeessarily origlnate from that area. Cod 1s a mlgratory fish, but the Inflow of new water tnflucnees the whole Baltle proper and thercfore it should not matter if the cod originates from thc areas ~~st or east of Bornholm. But oceasional improvements of thc oxygen eonditions should not eause the observed largc inereasing of the eod stock. That should probably be eonneeted to the inereasing cutrophication of thc surfaee water, regardlcss If this i5 eaused by natural or anthropogenie proccsses. 5 Referenees: Fonselius, S. 1969: Hydrography of the Balt1e Deep Basins III. Flshery Board of Sweden, Series Hydrography, No 23. Fonsellus, S. 1986: On long-term variations of dissolved oxygen in the deep water of the Baltte Sea. Raltie Sea Monitoring Symposium, Tallinn, USSR. 10-15 March 1Q86. P~oceedlngs. (in print). Manuscript. Larsson, U., R. Elmgren and F. Wulff. 1q85: Eutrophication and the Baltie Sea: Causes and Consequenees, Ambio, Vol. 111:1. • Meyer, P.F. and K. Kalle. 1980: Die biologische Umsti~~ung der Ostsee in den letzten Jahrzehnten - eine Folge hydrographischer Wasserumschichtungen. Archiv fur Fisehereiwissenschaften. 2:1-9. Nehring, 0., G. Aertebjerg, A. Trozosinska, S. Fonselius, P. Tulkki, P. Alenlus, V. Tervo, M. Bannus, V. Astok and A.. K. Yurkovaskis • 1q86: First Periodic Assessment of the State of the Baltte Sea, Chapt. 2. Nutrients. Balt1e Marine Environment Proceedings. (in print)~ t1anuscript. . Otterlind, G. 1986: On Fluctuatlons of the Baltte cod stock. J. Cons. Int. Explor. mer. (in print). Manuscript. Sehluz, S. 1985: " Ergebnisse ökologischer Untersuchungen 1m pelagischen Ökosystem der Ostsee, Akad. der Wissenseh, der DDR Inst. fur Meeresk. Rostoek-Warnemunde. Printed manuseript. 185 pp. Svansson, A.. 1985: Fiskerihydrografl. Medd. Havsfiskelab. Lysekil, Nr 307. Inst. Hydrogr. Res. Götehorg Series, No 31, 121 pp. Wulff, F., G. Aertebjerg, G. Nicolaus and A. Niemi. 1986: First perlodle Assessment of the State of the Baltlc Sea. Chapt. 4. Pelagic Biology. Baltle Marine Environment Proceedings. (in prlnt). Manuserlpt. Wyrtkl, K. 1954: Der gros se Salzeinbruch in die Ostsee 1m November und Dezember 1951. Kieler Meeresforschungen, 10:1. ... 0- ~~-==i:,==~~==d===i:,==~oh"'::x:;;=:c:i::t:"'=l -" 0:il o~ 0. 0:;; '" e: '" o~ 01: ->- Ln m ... • °0 U'l >- g) • o:!: 0", o::! 0;;; • Fig~ 1 Oensity distribution in the Gotland Oeep trom 1893 - 1985 0r------------------------------------------,oo ~ lllO~-_9---vrJ , 1 ~.2.\ 200 , I \1 " 8 Fig.2 OXYGEN CONOITIONS IN THE DEEP WATER IN THE GOTL~ OEEP FROM 1893 -1985 i: rE AR. Fig.3 0- •• u- • 3 mut )-2 2-1 . ·· '-0 · HaS "S~,' -::~ " . St " 15 A f1'I '0 11' n n Th. pIloJP/IGl. variolions in )'801 "11 " 'S " n 71 ". 10 11 "' . " os Fig.4 100 "" .. .. 10 se. I\' "" 11 n TI. n1&1.ate ....... 11 n J'f" Nil ,~ 11 Fig.S ,. GOTLANO BA51N o • --CD tS60-t985 SaL,-m.lu - ueer Ly means '\Qm r----.,..---r--~--r_--..-- __r__--r--_,.--..---., ,,-... '\0lIl -.J 8: CD ::n' -.Jr.J C .J ..JU) o • (flr- . .... 'Fi g.6 r- r = O.82~x ~'---_-'-_-.:.L _ _"""'_"""_ _" _ _ _ - & "_ '960 • '965 '910 GOTLANO BASIN __ PO,\P - ~ __'_ _- " - _ - - ' " _ - - - ' '915 1985 '980 Yaar t960-t985 yearLy meons '\0111 or---..,....---r---r----r--T-"-.,...---.,..---r--~-_, ..J ....-.... "-.J '\0lIl o Cl -- . jU) 0.. -r 0 o 0.. .... Fig.7 o " r:; O.78 Ol_._...&.._.-.I_ _......._.....L_ _"___-J.._--''---_..J__ 0'960 1965 '910 1915 1980 Jt)Cx _'"_--J 1985 Yaar GOTLANDBASIN t960-tS3S N03N - yeerLu meons '\Om ~ ._tnr----.----,--.,.....---r--,.--.,...---,r----r----..----, ..J ,,-... "- 'tOM -.JII) o . O)N j Zo N). ON Z U! - Fig.8 ~ ~l-...:;;.-'--~---:;.-'----J.--"---..J----'_-..J-_-'"---l 01960 4965 4910 4915 1990 4985 Yaar • , - lDOO t ann ; ~ .., \1 Total catch ot' cocl n the Saltic 1921-1964 ~ JlIO lDO lDO 19100 W50 t1. 1MO t t W70 3. 2. Fig.·9 80RNHCllM OEEP r 4- (Otterlind 1906) 8OtTOM SALINITY t~7-'9I2 (Fonse1ius et a1 1964, RPV no 1BS) " •" ~ \ :. (Svansson 1985) tod yeer class Good . Very good Bad Very bad + t Fi g.' 0