Chapter 9

Eutrophication, primary production and vertical export

10000

1000 ) -1 y -2

s m 100 e l o mm 10 Freshwater wetlands and lakes P inputs ( Forests Paul Wassmann 1 Agroecosystems Estuaries and Norwegian College of Fishery Science coastal waters University of Tromsø 0.1 N-9037 Tromsø, NORWAY 10 100 1000 10000 10000 N inputs (mmoles m-2 y-1) [email protected] Figure 9.1: Nitrogen and phosphorous loadings to differ- ent types of ecosystems (redrawn from Eos, 1992). 9.1 Introduction

The effects of global and local changes are most across the land-sea margins at such high rates that prominent at the land-sea margins where presently coastal waters and estuaries are the most fertilized population growth is greatest. For example, the ecosystems on earth (Figure 9.1). population of coastal counties of the USA has The discharge of nutrients to the coastal zone roughly doubled since 1960 (Eos, 1992). This gives increased strongly during the last centennial. The rise to increased pressure on natural resources increase in human population, the use of fertilis- and a large number of disturbances to coastal re- ers, increased intensity in agriculture, logging and gions. Presently, eutrophication of coastal wa- increased atmospheric deposition are the main ters is probably the most important environmen- cause for this intensification. However, signifi- tal effect (Gesamp, 1991). The effects of nutrient cant periods of eutrophication took place much enrichment thoroughly change coastal ecosystems earlier. Already during pre-Roman time signifi- and occur virtually worldwide. Nutrients move cant amounts of the mixed woods in the Mediter-

126 9.3. PRIMARY PRODUCTION AND VERTICAL EXPORT 127 ranean disappeared. During pre-Viking times nat- nutrient resources (Richardson & Jørgensen, ural woods disappeared in Denmark, during the 1996) Middle Ages most of the original woods from cen- tral and northern Europe. Europe developed into 2. An increase in the rate of organic carbon pro- the cultural landscape that we encounter today, duction in an ecosystem (Nixon, 1995) unrecognisable and widely different compared to the pristine state. The removal of woods and In most cases we use definition 1 and this will the introduction of agriculture had a strong im- also be the case in the present text. Thus we fo- pact on the leaching of nutrients all over Europe cus upon the supply to and the dynamics of nu- and periodically eutrophication in the Baltic and trient resources in a water body. Eutrophication North Seas must have been significant over the can entail either the process or the result. One has last 1000 years. Cultural eutrophication, intense further to distinguish between natural eutrophica- as it may be at present, is thus nothing new, but a tion that is caused by winter accumulation, pre- close and well-known companion of human civili- cipitation, vertical mixing, upwelling, river run-off sation, mainly through the introduction of agricul- and entrainment of nutrients. Climatic variability ture that paved the road for human development obviously influences and modulates the nutrient and population explosion. availability of a recipient and natural eutrophi- Coastal ecosystems can accommodate large cation thus varies over time. The natural vari- amounts of nutrients, but there is certainly the ability in eutrophication is often poorly known, danger that increased loading gives rise to in- in particular because it may be camouflaged by creases in suspended biomass, far beyond the cultural eutrophication, which is any type of nu- range of natural concentrations. For scientists trient discharge caused by anthropogenic activity, and managers alike the question arises how much e.g agriculture run-off (see Sections 3), sewage, at- nutrient discharge a specific recipient can accom- mospheric deposition (see Sections 1), changes in modate per unit time before undesirable conse- water discharge etc. quences occur. Can we determine the primary production rates where excessive biogenic matter 9.3 Primary production and is exported to the bottom water and the sedi- ment that result in bottom fauna changes and ul- vertical export: Background timately anoxia? considerations

Primary production consists of new production 9.2 Eutrophication (PN) that is based on allochthonous, i.e. exter- nally supplied nutrients, and regenerated produc- The term eutrophication derives from the Greek tion (PR), which is based on autochthonous, i.e. roots eu (‘well’) and trope (‘nourishment’) and internally recycled nutrients (Dugdale & Goering, could thus be translated into well-fed, well- 1967). Hence total primary production (PT) is the nourished. With the term eutrophication we im- sum of PN and PR. The amount of carbon that ply that the ecosystem, not an individual, is well- enters the aphotic zone is entitled export produc- nourished and that nutrients or biomass are sup- tion (PE) (Figure 9.2). plied to a particular recipient. Eutrophication is The concept of new production is of utmost im- not a clearly defined term and there are various portance for understanding natural and eutrophi- definition such as: cated ecosystems because the fraction f = PN /PT represents the upper limit of organic matter and 1. The process of changing the nutritional sta- energy which can be removed or extracted from tus of a given water body by increasing the the surface waters of the system without de- 128 CHAPTER 9. EUTROPHICATION, PRIMARY PRODUCTION AND VERTICAL EXPORT

Given the practical difficulties in estimating P Total primary production N for lengthy periods of time, sediment traps can be used to estimate PN.PE estimates as measured by

e sediment traps come close to PN, but are always n o New Regenerated z smaller because it comprises only the particulate production tic production ho fraction and some transformation takes place from p u E ammonium to nitrate even in the upper layers. Calculations of the productivity index f by ap- plying PE give, therefore, rise to underestimates Export production (Wassmann, 1993). As a consequence, the term (Sedimentation) e = PE/PT can be applied and used as an approx- imation of f. In boreal, coastal areas where steady Figure 9.2: New and regenerated production are based on (a) the supply of the limiting (allochthonous) nutri- state, if at all, cannot be assumed for intervals of ents from the aphotic zone, by advection, run-off or from less than the length of the productive period, e is the atmosphere (straight arrows) and (b) the recycled (au- meaningful as a base for estimating f for lengthy tochthonous) nutrients in the euphotic zone (circular ar- periods only (e.g. > 6 months). Therefore, the rows), respectively. New and regenerated production com- prise total primary production. Export production is the term < e >, representing e for lengthy periods of amount of sinking organic carbon at the bottom of the eu- time will be applied. photic zone. 9.4 Nutrient supply, primary stroying the long-term integrity of pelagic sys- production, retention and tems (Vezina & Platt, 1987; Iverson, 1990; Legen- vertical export dre, 1990). PN represents thus the biomass that has to be handled by an eutrophicated recipient Increased supply of nutrients to the euphotic zone (e.g. mineralization, accumulation, harvestable gives rise to increased production of algae which biomass and export of biomass to adjacent recip- sooner or later sink to deeper water and the sed- ients). Given the importance of PN for the over- iment, resulting in increased sediment-water ex- all cycling of organic matter, considerable empha- change rates, at times in mass mortality of macro- sis has recently been given to estimating PN in fauna and fish eggs (Rosenberg & Loo, 1988; Mor- coastal (e.g. Wassmann, 1990b) as well as oceanic rison et al., 1991) and finally in anoxia (Rosenberg, environments (e.g. Knauer et al. 1990). New 1985; Graf, 1987). During the last decades wide- production represents the carrying capacity of a spread occurrence of low oxygen concentrations marine ecosystem. New production represents the or anoxia in bottom waters, decreased catches of maximum production capacity of an ecosystem or fish and blooms of toxic algae threatening aqua- the harvestable production. New production is culture as well as stocks of wild animals have been a critical component of marine primary produc- reported with increasing frequency (Rosenberg & tion that limits the supply of food to the ben- Loo, 1988). These changes seem to be caused by thos, zooplankton, fish, and extensive aquaculture increased inputs of nutrients to aquatic areas from as well as the removal rate of atmospheric CO2 sewage, agricultural run-off and atmospheric fall- by the marine biota. New production estimates out, giving rise to various degrees of eutrophica- are of great interest for understanding eutrophi- tion of fresh-water as well as marine, coastal envi- cation. Increased new production results in addi- ronments (e.g. Wulff et al., 1990). tional biomass that the ecosystem has to deal with Figure 9.3 shows the principle processes of at- in terms of grazing, vertical export to the bottom mospheric CO2 uptake and release, primary pro- and pelagic and benthic degradation. duction, suspended biomass and vertical export to 9.4. NUTRIENT SUPPLY, PRIMARY PRODUCTION, RETENTION AND EXPORT 129 the bottom. Seawater takes up CO2 from the at- mosphere that is either taken up by phytoplankton or released again to the atmosphere. The phyto- plankton uptake of CO2 is caused by primary pro- duction and first and foremost dependent on pho- tosynthetic active radiation and nutrients. In ad- dition it is influenced by the residence time of phy- toplankton in the euphotic layer (determined by vertical mixing and stratification). Phytoplank- ton accumulates in the upper layers if grazing and degradation rates are lower than primary produc- tion, i.e. a bloom takes shape. A part of the CO2 suspended biomass, consisting of phytoplankton cells and detritus will inevitably escape grazing CO2 Phytoplankton and degradation and sink into the aphotic zone Euphotic and further to the bottom. The export of bio- Zoo- zone genic matter to the bottom is a complex func- oduction plankton New pr Regeneratedoduction tion of the total amount of suspended matter, the New pr Fish nutrients sinking velocity of the sinking particles and the Higher Recycled degradation impact of the pelagic heterotrophs. nutrients trophic levels For example, low suspended biomass, low pelagic degradation and high sinking rates give result in Sedimentation (Export production) a similar sedimentation rate at depth than high Grazing suspended biomass, high pelagic degradation and Microbial degradation Aphotic low sinking rates. Degradation of organic matter Dissolution zone in the water column or the sediment results in nu- Release of trients that sooner or later can be taken up by nutrients phytoplankton. If the limiting nutrient is nitrogen the new pri- mary production depends on the allochthonous nutrient nitrate while the remaining primary pro- duction is based upon the autochthonous nutrient ammonium that derives from internal recycling by Figure 9.3: Primary production, vertical flux and regen- heterotrophic organisms. The basic principle to eration of nutrients in a coastal marine ecosystem. Also use nitrogen species to determine how much of the shown are some of the involved organisms such as phyto- total primary production comprises new produc- plankton, zooplankton, higher trophic levels and benthic organisms. The massive and narrow vertical arrows indi- tion (in case nitrogen is the limiting element) is cate scenarios of substantial and insignificant vertical flux. difficult to apply in shallow water where the cycle (Illustration courtesy: dr. Alexander Keck.) of nutrients is rapid and where particulate nitro- gen supplied to the bottom can be recycled to ni- trate that is available for primary production. In this case nitrate is not new, but regenerated. Thus some of the nitrate is not ‘new’ and does not com- ply with the basic assumption of new production. In countless eutrophicated regions ammonium and urea are supplied as allochthonous nutrients. Also 130 CHAPTER 9. EUTROPHICATION, PRIMARY PRODUCTION AND VERTICAL EXPORT in this case the traditional method to distinguish 120 ) new from regenerated production is not possible -1

y 100 because some of the per definition autochthonous -2 m

C

nutrients are allochthonous. New production is g 80 ( n o thus impossible to measure in eutrophicated wa- i t c ters. u 60 d o The state of an ecosystem during a transient r t p 40 or

bloom is basically characterised by export food p Ex chains with high vertical export. The amount of 20 regenerated production increases, as the plank- tonic system develops and becomes more com- 40 80 120 160 200 240 plex during the post bloom phase. Sedimenta- Total primary production (g C m-2 y-1) tion of organic material is low and the ecosystem is characterised by retention food chains. In the Figure 9.4: Export production as a function of total pri- non-eutrophicated coastal zone export chains are mary production on an annual scale in marine ecosystems. Algorithms from various publications are presented. Suess based upon new production and represent episodic (1980) (...... ), Eppley & Petersen (1979) (- - - - -), Bet- events on the background of a continuous, season- zer et al. (1984) (-.-.-.-.-), Pace et al. (1987) (-..-..-..-) and ally variable regenerated production based on the Wassmann (1990) (—). Source: Wassmann (1990b, 1993). recycled nutrients from retention chains. If eu- trophication continues, i.e. nutrients are supplied ships in the various ecosystems from which they in a steady manner, a new steady state with a were derived (Figure 9.5)? If so, then different mixture of export and retention food chains will algorithms should be applied in different regions. develop. In particular data from the boreal coastal zone In conclusion, an estimate of from the North Atlantic were investigated. The new/net/harvestable production as a conse- data used was mainly selected from simultaneous, quence of eutrophication in coastal zones is time-integrated measurements derived over inter- difficult to measure, among other reason because vals covering most of the productive season (>6 our terminology and measuring techniques are months). Through a regression analysis PE was inadequate. positively and nonlinearly correlated with total 2 production PT (Figure 9.5). Best fit (r = 0.94) 9.5 Algorithms of primary pro- was found by a power model calculated by the duction versus vertical car- equation: bon export 1.41 PE = 0.049PT (9.1) An overview on algorithms predicting export pro- The < e > ratio was also calculated and both duction on the base of total primary production < e > and PR were found to be positively, non- in marine environments on an annual scale has linearly correlated with PT. The upper limit for been presented by Wassmann (1990b; 1993) (Fig- < f > was calculated to be about 0.5 in boreal ure 9.4). Significant variability with regard to the coastal environments, i.e. at most about 50% of PE versus PT relationship was detected. What al- PT may be exported through sedimentation to be- gorithm should be selected for a global or coastal low the euphotic zone. The curvilinear nature of eutrophication carbon flux model? Obviously, the relationship implies that vertical export of bio- there is no universal algorithm that would fit all genic matter increases relatively more than total ecosystems. Does the variability of the algorithms primary production. reflect real difference in the PE vs. PT relation- The results of the model of Aksnes & Wass- 9.5. PRIMARY VS. EXPORT PRODUCTION ALGORITHMS 131

mann (1993) indicate that domination by cope- pods in the marine and cladocerans in lakes can give rise to very different relationships between primary versus export production (Figure 9.5). Meso-zooplankton species composition obviously influences the pelagic-benthic coupling: for exam- 1 ple, copepods and cladocerans have different re- ) 1 - 140

y A productive strategies (hence different grazing pres- 2 - 120 m sure), and cladocerans do not produce distinct fae- 100 (g C

cal pellets. A comparison of retention and export 2 on

i 80 t food chains, and vertical flux in lakes dominated c u

d 60 by copepods (e.g. Lake Baikal) or marine envi- o 40 ronments strongly influenced by cladocerans (e.g. t pr

or 20 the eastern Baltic Sea), would be advantageous to Exp analyse in greater detail the contrasting scenarios )

1 of copepod and cladoceran dominance for pelagic- - 140 B y 2 - 120 benthic coupling. m In case the algorithms depicted in Figure 9.5 100 (g C are truly predicting annual PE on the base of on

i 80 t

c PT, why are there significant differences? In the u

d 60

o case of subalpine lakes and boreal coastal areas 40 t pr we have already recognised that differences in the or 20 zooplankton community species composition re- Exp 100 200 300 400 500 sult in the observed variance. The question can be Primary production (g m-2 y-1) raised if the results presented in Figures 9.4 and 9.6 suggest that various types of top-down reg- Figure 9.5: (A) Export production as a function of total ulation are the base for the observed variability? primary production from the North Atlantic, boreal coast (Wassmann, 1990; full line) and subalpine lakes (Aksnes The few data which do exist from non-boreal envi- & Wassmann, 1993; broken line) on an annual scale. The ronments outside the North Atlantic suggest that zooplankton of the former ecosystems is often dominated coastal areas and tropical bays in the North Pacific by copepods, the latter one by cladocerans. Also shown Ocean experience more efficient retention in the are two data points from Dabob Bay, a boreal, North Pa- cific fjord, and a tropical lagoon, Kaneohe Bay on Hawaii upper layers and less vertical export (Figure 9.5). (open squares, 1 and 2, respectively). (B) Schematic di- This interpretation is in consistency with the no- agram on the conceivable relationship between annual ex- tion that tropical environments are characterised port production and total primary production in miscella- by effective retention food chains. This may also neous ecosystems with different production, recycling and export regimes. The functional lines of the various ecosys- be true for the North Pacific Ocean where at tems could be spread in the shaded area. The relationships least the open ocean is characterised by extensive could fall onto a suite of lines contrasting between maxi- micro-zooplankton grazing which prevents major mum export (steep angle, straight relationship) and high accumulation of phytoplankton biomass (Frost, retention (flat angle, strong curvature) efficiencies. 1991; Dagg, 1993). PE as a function of PT in miscellaneous ecosystems with different produc- tion, recycling and export regimes could fall onto a suite of lines falling between maximum export (steep angle, straight line = bottom-up regula- tion) and high retention (flat angle, curved line = top-down regulation) efficiencies (Figure 9.6). The 132 CHAPTER 9. EUTROPHICATION, PRIMARY PRODUCTION AND VERTICAL EXPORT

Bottom-up dicted by Wassmann (1998). Increased bottom- regulation up regulation by eutrophication will increase the ‘loopy’ nature of the PT vs. PE relationship. In contrast, increased top-down regulation will de-

t crease the loop size. Increased top-down regula- or

p tion will eventually force the loop onto a retention x e

l line and remove excessive vertical export. a c ti r e V 9.6 Increases in primary and ex-

Primary productino (m-2 y-1) port production: Examples from the Gullmaren Fjord Figure 9.6: Schematic representation of annual primary production and vertical export during a phytoplankton and the Kattegat bloom. Increased new production drives the relationship along the linear relationship total primary production = Two cases studies illustrate that a ‘threshold in- new production = export production. Planktonic het- erotrophs reduce vertical export. As the grazing capac- terval’ in primary production exists where vertical ity of the planktonic heterotrophs increases with increasing export increases strongly. The pelagic ecosystem primary production (full arrows), a curvilinear relationship of the Gullmar fjord situated on the west coast emerges. Note the horizontal bars that indicate threshold of Sweden and adjacent waters has been stud- intervals where the curvature of the primary production vs. ied since the late 1970s, principally in relation to vertical export relationship increases rapidly. Carnivory (stippled arrows) counteracts the retention of suspended oceanographic variability in the Skagerrak and the biomass by the herbivores and detrivores. possible influence of climatic forcing on this area (Lindahl and Hernroth, 1983; Andersson and Ry- dberg, 1993; Heilmann et al., 1994; Lindahl et al., balance between bottom-up and top-down regula- 1998; Belgrano et al., 1999). Primary phytoplank- tion shapes the curvilinear nature of the PT vs. ton productivity has been a part of these studies PE relationship. Not only PT varies as a function and a measuring program in the mouth area of of climate variability and eutrophication, also the the Gullmar Fjord is ongoing since 1985. An eval- PT /PE ratio is not constant, but varies in accor- uation of this time series was carried out in 1994 dance with the composition and dynamics of the (Lindahl, 1995), suggesting that even when ele- heterotrophic plankton community. ‘The’ PT vs. vated values of primary production are observed PE relationship does that not exist. during the spring period (March-April), the main On a daily scale the PT vs. PE relationship is contribution to the annual production was found characterised by irregularities (Figure 9.7). Pri- during the period May-September. mary production varies greatly between days and More recently a first attempt was carried out contributes to in different degrees to the sus- to study the effect of weather/climatic forcing on pended pool of biogenic matter (a function of total the physical-chemical processes related to the pri- production, the f-ratio, grazing etc.) that may mary productivity. These results suggested the sink. Although short-term variability in vertical presence of an indirect link between the North flux takes place, that of primary production is Atlantic Oscillation index (NAO), the supply of greater. The phase plot in Figure 9.7 illustrates nutrients to Kattegat, wind direction and the pri- the spiky nature of primary production as com- mary production (Lindahl et al., 1998; Belgrano pared to the buffered rates of vertical export. The et al., 1999). The development of primary pro- 5 days running average plot indicates the loop- duction was reconstructed by combining measure- type relationship between daily PT and PE, as pre- ments in the Gullmaren fjord with older measure- 9.6. GULLMAREN FJORD AND KATTEGAT EXAMPLES 133

0.5 0.5 Polar Front 0.4 0.4 t 0.3 0.3 Expor 0.2 0.2

Figure 9.7: Unpublished results 0.1 0.1 from a physically-biologically cou- Atlantic Year: 1999 pled 3D model presenting primary 0 0 production versus export production 0 0.5 1 1.5 2 0 0.5 1 1.5 2 in 4 different regions in the Barents 0.5 0.5 Arctic south Arctic north Sea (pers. com., D. Slagstad). The 0.4 0.4 scattered line is the daily variability t in the phase diagram while the loops 0.3 0.3 are the 5 day running average. Expor 0.2 0.2

0.1 0.1

0 0 0 0.5 1 1.5 2 0 0.5 1 1.5 2 Primary prduction Primary prduction ments from the Kattegat (Figure 9.8). The strik- ing increase in the 70s and 80s seems caused by Development of primary production eutrophication, while the slight rise is interpreted in the Gullmaren fjord as a function of climate change. Applying the rela- ) -1 300 tionship suggested by Wassmann (1990b), the ex- year 250 Increased -2 port production in the 1950/60 period was about 200 eutrophication -2 -1 -2 -1 30 g C m year (PT = 100 g C m year ) while 150 -2 -1 Increase due to at present it is about 120 g C m year (PT = 240 100 climate change -2 -1 g C m year ). If the assumptions behind these 50 Original level

calculations are true, they imply the vertical C ex- tical export (g C m 0 r e port increased four times over a time interval of V 1940 1960 1980 2000 50 years! The carbon loading of the basin water of Vertical export the fjord is obviously far greater today then during Figure 9.8: Development of primary production in the the more ‘pristine’ times prior to 1960. Gullmaren fjord over the last 30 years. But even over the recent period significant in- creases in the organic load to the deep part of change over time in deep-water exchange. Finally, Gullmaren Fjord below the euphotic zone can be it should be mentioned that the relationship be- calculated. PE has increased from approximately tween PT and PE in the fjord reflects both eu- 105 g C m-2 y-1 in 1985 to almost 123 g C m-2 trophication (which has not increased significantly y-1 in 2000, corresponding to an increase of the in recent years due to increased effluent control) organic load of about 17% over 15 years. One and climate changes (variations in NAO, global possible result of this process may be the observed warming etc.). However, to differentiate between decrease in oxygen content of the deep water (>60 natural and anthropogenic variability is difficult. m) the beginning of the 1980s. However, it should Applying again the PT and PE relationship of be pointed out that the decrease in oxygen may Wassmann (1990b), the vertical export from the be explained by other processes as well, e.g. a upper layers in the Kattegat region appears to 134 CHAPTER 9. EUTROPHICATION, PRIMARY PRODUCTION AND VERTICAL EXPORT

Table 9.1: Increase in primary production in the southern Kattegat over time. Also shown the calculated change in export production (from Wassmann, 1990a).

Area Time interval Change PT Change PE covered (years) (g C m-2 y-1) (g C m-2 y-1) Storebelt 24 +63 +29 Øresund 44 +58 +22

have increased 130–250% over a time interval of 20–40 years (Table 9.1; Wassmann, 1990a). This fundamental increase should be adequate to ex- plain the frequently observed oxygen deficiencies in the region (Rosenberg & Loo, 1988), although, Vertical flux as mentioned above, stratification and lack of ex- change of bottom water results in anoxic condi- tions. The curvilinear nature of the PT versus PE relationship implies that the linear increase in PT causes an exponential type of increase in PE, -2 -1 in particular at PT rates >150 g C m y . It would be advantageous to determine the primary th

production threshold intervals for various coastal p e regions where PE turns out to be greater then oxy- D gen content of or supply to the benthic bound- ary layer and where undesirable effects (hypoxia, anoxia) develop in the bottom layers.

9.7 Variability of vertical export in the pelagic zone

All investigations of export of biogenic matter in- Figure 9.9: Schematic presentation of the ‘pelagic mill’ dicate that the export flux decreases more or less in the upper part of the ocean and its regulation of biogenic exponentially with depth in the upper part of vertical flux. The full line assumes a continuous minerali- sation of export production, giving raise to a decline in flux the ocean, with minor decreases below 200–500 m that follows a power function. The broken line indicates depth (for algorithms predicting the depth varia- a step-wise decrease in vertical flux caused by extensive tion of vertical carbon flux see Berger et al. 1989). grazing at certain depth horizons. The stippled line indi- Resuspension and protrusion of advective, particle cates that vertical flux can increase intermittently due to rich layers or vertical differences in current direc- repackaging. The recycling by the zooplankton community is schematically indicated to the right. tion may alter this general feature of vertical flux. The degradation rate of organic matter in the wa- ter column and, in particular for fast sinking par- ticles is of pivotal importance for the quantitative regulation of pelagic-benthic coupling. Depend- 9.8. SEASONAL VARIATION IN VERTICAL EXPORT 135

ing on the degradation rate of fast sinking par- 600 ) -1 ticles in teh water column, the absolute vertical y a d export of organic matter at a certain depth could -2 m 400 be small or large, irrespective the size of the new g C m ( production from which it derives. n tio a t

The current lack of adequate investigations of n e 200 m i the vertical export above the depth of 200–500 m d e where the majority of long-term sediment traps S have been deployed, results in difficulties to un- F M A M J J A S derstand and model vertical carbon flux. There Time (months) exists a black box of several hundred metres be- Figure 9.10: Semiquantitative diagram of annual varia- tween the surface layers where measurements and tion of sedimentation of particulate organic carbon (POC) algorithms of primary production exists and where in fjords, eutrophicated as well as non-eutrophicated polls. data on the carbon export to the ocean interior are One-, two- or multi-pulse systems can be distinguished. The characteristic line for eutrophicated fjords is the dotted available. In this black box, the twilight zone, we line (increased average, reduced seasonal variability). face a lack of basic understanding on how vertical export of biogenic matter in general is regulated, let alone in eutrophic regions. In order to guide matter is large, but the average sedimentation rate future investigations of vertical flux attenuation is low. in eutrophicated regions we present an idealised, Eutrophicated land-locked fjords show also sea- conceptual model of vertical carbon export and sonal variability in organic matter flux, but the focus upon the ‘pelagic mill’ and vertical flux reg- relative amplitude of new production, suspended ulation in the upper 200 m (Figures 9.3 and 9.9). matter and sedimentation is lower (Figure 9.10). An adequate understanding of carbon cycling de- The average sedimentation rate, however, is high. mands not only adequate investigations of primary The variability in such ecosystems is ‘buffered’ by production, but also concomitant research on the the continuous supply of allochthonous nutrients. functional biodiversity of the pelagic zone, plank- Eutrophicated land-locked fjords are mainly one- ton dynamics, vertical flux and its regulation in pulse systems, but depending on the supply of nu- the twilight zone. trients from fresh water run-off, sewage etc. and climatic conditions, several minor summer and au- tumn blooms may develop. They may thus turn 9.8 Seasonal variation in verti- into two- or multi-pulse systems (Figure 9.10). cal export in eutrophicated Open fjords are complicated multi-pulse sys- coastal areas tems. Pulses in spring and autumn are normally found. However, upwelling of nutrient rich deep Considering the seasonal flux of organic matter water can introduce additional pulses to the sys- in various coastal settings three major modes can tem at any time, but normally during late spring be distinguished: one-pulse, multi-pulse and in and early summer (Figure 9.10). Also, accumu- eutrophicated regions ‘buffered’ systems (Wass- lated biomass can be removed from the fjord by mann, 1991). As an example, we present data large-scale exchange of water. In multi-pulse sys- from west-Norwegian fjords (Figure 9.10). Non- tems advection represents the most significant el- eutrophicated land-locked fjords represent simple ement. one-pulse systems, where new production, sus- Comparing the dynamics of primary production pended biomass and sedimentation give rise to one and sedimentation in fjords renders, therefore, dif- major, annual pulse during spring. The relative ficult because of the differences in time and space amplitude of the seasonal signal in flux of organic scales of these processes. 136 CHAPTER 9. EUTROPHICATION, PRIMARY PRODUCTION AND VERTICAL EXPORT

Primary production is usually estimated in ton and vertical flux of biogenic matter as pulsed terms of litres and hours, sedimentation, how- nutrient addition resulted in the highest vertical ever, integrates the vertical flux at a given export. depth over the time of trap deployment and is expressed in terms of square meters and days. While the produced biomass can stay 9.9 Eutrophication and phyto- in the fjord or is dispersed in adjacent bod- plankton biomass accumula- ies of water, sediment traps might catch or- tion ganic particles that have been produced and altered throughout the coastal zone. If ad- The influence of top-down control is obviously im- vection is significant in fjords, the locally portant for the flow of nutrients through the food measured primary production and sedimen- chain or food web. In lakes the cascading effects tation rates might have little in common, through the food web by manipulating the top- but rather reflect the general productivity down regulation is well known (Mazumder et al., and vertical flux regime in all parts of the 1988). Top-down effects have less-known effects coastal zones, from the innermost reaches to on marine coastal eutrophication. As most of the the open shelf. This has also implications for eutrophicated regions are in the shallow coastal eutrophication. Eutrophication-derived sus- zone some of the peculiarities of these ecosys- pended biomass may be introduced into a tems have become mixed up into the term eu- non-polluted region from outside or local eu- trophication that is almost analogue with green trophication signals may be exported to un- or brown waters. Do green or brown waters in- polluted regions. dicate marked increases in marine productivity or eutrophication? In most cases in the coastal Ecosystems that receive nutrients continuously zone this is the case, but it could also just re- or pulsed differ with regard to the pelagic-benthic flect the lack of important grazers such as cope- coupling. Pulsed nutrient addition may cause pods that are excluded from overwintering in shal- a higher build-up of phytoplankton biomass, a low waters, resulting in decreased grazing pressure larger temporal mismatch between herbivores and on large-celled bloom phytoplankton. Primarily phytoplankton biomass and a higher sedimenta- brown and green waters do not suggest that there tion rate of biogenic matter. This was tested in is less grazing than phytoplankton production: a enclosures (Svensen et al., 2002). Each enclosure mismatch between producers and consumers. Do received the same total amount of nutrients, but blue water indicate that marine productivity is the nutrients were supplied at four different inter- low? In some cases this is true, in others not. vals ranging from one single load to continuous Blue water reflects a balance between producers additions. Spring bloom-like systems developed and consumers: biomass accumulation does not where nutrients were added in one or two pulses as take place. Production could be high or low. Con- they were characterised by high primary produc- tradicting to common believe, some blue waters tion, high suspended biomass of chlorophyll a (Chl are highly eutrophic (e.g. the north Norwegian a) and particulate organic carbon (POC) and high shelf; Wassmann et al., 1999) while others are olig- sedimentation rates. In contrast, the seawater en- otrophic (e.g. the central and eastern Mediter- closures receiving nutrients about every third day ranean Sea). These findings have obvious impli- or in a continuous supply resembled regenerated cations for the interpretation of pigment data and systems with low concentrations of suspended Chl remotely sensed pigments concentrations that tra- a and POC and with low and stable loss rates. ditionally have been applied to construct PT fields The frequency of nutrient additions had a strong over regions where PT measurements were unavail- influence on the development of the phytoplank- able. There cannot exist a constant pigment/PT REFERENCES 137 ratio, analogous to that no constant PT/PE ratio system in the coastal northern Baltic Sea. Marine Ecol- exists. The top-down regulation of phytoplank- ogy Progress Series, 145, 195–208. ton biomass is thus important to keep in mind Iverson, R. L. 1990. Control of marine fish production. when the eutrophic status of a region is estab- Limnology and Oceanography, 35, 1593–1604. lished. Blue water can produce large amounts of Knauer, G. A., Redalje, D. A., Harrison, W. G., & detritus and result in large-scale vertical export of Karl, D. M. 1990. New production at the VERTEX biogenic matter, resulting in large supply to the time series site. Deep Sea Research, 37, 1121–1134. benthos and oxygen deficiency in bottom waters. Legendre, L. 1990. The significance of microalgal blooms The cascading effect of top-down manipulation for fisheries and for the export of particulate organic car- bon in the ocean. Journal of Plankton Research, 12, influenced the plankton community and results in 681–699. different functional response in the various regions exposed to eutrophication. During the process Mazumder, A., McQueen, D. J., Taylor, W. D., & S., Keabm D. R. 1988. Effects of fertilisation and planktiv- of eutrophication, the food web structure, timing orous fish (yellow perch) predation on size distribution of fertilisation and alternative grazing/predation of particulate phosphorus and assimilated phosphate: strategies of the planktonic heterotrophs have a Large enclosure experiments. Limnology and Oceanog- crucial impact on the retention and loss of nutri- raphy, 33, 421–430. ents from the pelagic zone (Heiskanen et al., 1996; Morrison, J. A., Napier, J. R., & Gamble, J. C. 1991. Svensen et al., 2002). Mass mortality of herring eggs associated with a sedi- menting diatom bloom. ICES Journal of Marine Sci- ence, 48, 237–245. References Nixon, S. W. 1995. Coastal marine eutrophication: a def- Aksnes, D. L., & Wassmann, P. 1993. Modelling the sig- inition, social causes, and future concerns. Ophelia, 41, nificance of zooplankton grazing for export production. 199–219. Limnology and Oceanography, 38, 978–985. Richardson, K., & Jørgensen, B. B. 1996. Eutrophica- Berger, M. H., Smetacek, V. S., & Weger, G. 1989. tion: Definition, history and effects. Pages 1–20 of: Eu- Ocean productivity and paleoproductivity - an overview. trophication in coastal marine ecosystems. Coastal and Pages 1–34 of: Berger, W. H., Smetacek, V. S., & Esturarine Studies, vol. 52. Washington DC: American Wefer, G. (eds), Productivity of the ocean: present and Geophysical Union. past. New York: John Wiley & Sons. Rosenberg, R. 1985. Eutrophication — the future marine Dagg, M. 1993. Grazing by the copepod community does coastal nuisance? Marine Pollution Bulletin, 16, 227– not control phytoplankton production in the Subarctic 231. Pacific Ocean. Progress in Oceanography, 32, 163–183. 1988. Marine eutrophica- Dugdale, R. C., & Goering, J. J. 1967. Uptake of new Rosenberg, R., & Loo, L. O. and regenerated forms of nitrogen in primary productiv- tion induced oxygen deficinecy: effects on soft bottom ity. Limnology and Oceanography, 12, 196–206. fauna, western Sweden. Ophelia, 29, 213–225. Eos. 1992. Understanding changes in coastal environmnets: Svensen, C., Nejstgaard, J. C., Egge, J. K., & Wass- the LMER Program. Eos, 73, 481–485. mann, P. 2002. Pulsing vs. constant supply of nu- trinets (N, P and Si): effect on phytoplankton commu- Frost, B. W. 1991. The role of grazing in nutrient-rich nity, mesozooplankton grazing and vertical flux of bio- areas of the open sea. Limnology and Oceanography, 36, genic matter. Scientia Marina, 66, 189–203. 1616–1630. Gesamp. 1991. The state of the marine environment. Ox- Vezina, A., & Platt, T. 1987. Small-scale variations ford: Blackwell. of new production and particulate fluxes in the ocean. Canadian Journal of Fishery and Aquatic Science, 44, Graf, G. 1987. Benthic response to annual sedimenta- 198–205. tion pattern. Pages 84–91 of: Ruhmor, J., Walgert, E., & Zeitzschel, B. (eds), Seawater-Sediment Inter- Wassmann, P. 1990a. Calculating the load of organic car- actions in Coastal Waters. Lecture Notes on Coastal and bon to the aphotic zone in eutrophicated coastal waters. Estuarine Studies, vol. 13. Berlin: Springer Verlag. Marine Pollution Bulletin, 21, 183–187. Heiskanen, A.-S., Tamminen, T., & Gundersen, K. Wassmann, P. 1990b. Relationship between primary and 1996. The impact of planktonic food web structure on export production in the boreal coastal zone of the North nutrient retention and loss from a late summer pelagic Atlantic. Limnology and Oceanography, 35, 464–471. 138 REFERENCES

Wassmann, P. 1991. Dynamics of primary production and sedimentation in shallow fjords and polls of western Nor- way. Oceanography and Marine Biology Annual Review, 29, 87–164. Wassmann, P. 1993. Regulation of vertical export of particulate organic matter from the euphotic zone by planktonic heterotrophs in eutrophicated aquatic envi- ronments. Marine Pollution Bulletin, 26, 636–643. Wassmann, P. 1998. Retention versus export food chains: processes controlling sinking loss from marine pelagic en- vironment. Hydrobiologia, 363, 29–57. Wassmann, P., Andreassen, I., & Rey, F. 1999. Sea- sonal variation of nutrient and suspended biomass along a transect on Nordvestbanken, north Norwegian shelf, in 1994. Sarsia, 84, 199–212. Wulff, F., Stigebrandt, A., & Rahm, L. 1990. Nutri- ent dynamics of the Baltic Sea. Ambio, 19, 126–133.