The Role of the Microbial Food Web in Ecosystem-Based Management

ICES 2008 Theme Session A. ICES CM 2008/A:01

The role of the microbial food web in ecosystem-based management.

Michael R. Heath, Marine Laboratory, Aberdeen, UK

John H. Steele, Woods Hole Oceanogr. Instn., USA

Abstract

The present focus on ecosystem-based management (EBM) for fisheries has produced much work on budgets for nutrient or energy flow through ecosystems; usually with emphasis on the higher trophic levels. Some end-to-end studies use the ratios of fish yield to net primary production as a system index, but this ratio is very variable and can reflect differences in the factors determining nutrient recycling within the lower trophic levels, rather than stresses on the upper trophic components. We argue that explicit consideration of these physical, biogeochemical and ecological processes is essential if we are to understand the constraints on overall system productivity and the causes of changes in this productivity. We illustrate these issues with comparative analyses of the Georges Bank and the North Sea ecosystems.

Contact author: John Steele: Woods Hole Oceanographic Institution, Woods Hole MA 02543, USA. [Tel. 508 289 2220: Fax 508 457 2184: Email: [email protected].]

Introduction

The interest in ecosystem-based management (EBM) has created a need for a variety of end-to-end representations of marine food webs in order to explain the community production and diversity of harvestable trophic levels, rather than just the dynamics of individual species in isolation. One approach has been to empirically relate higher trophic level production to primary production, so that spatial and temporal patterns in primary production can then be used as a basis for assessing the maximum potential fishery yield. There is considerable spatially and temporally resolved information on rates of net or total annual primary production by phytoplankton (TPP) based on a combination of remotely sensed color data and experimental measures using tracer (14C) uptake during incubations. However, relationships between primary and fish production seem to vary by an order of magnitude, Table 1. This variability may be partly due to differences in the way fish production is calculated, but must also indicate large variability in the efficiency with which primary production is transmitted up the food web. Its is clear that use of this approach as a basis for EBM requires a better understanding of the food web structure and function.

There are a large number of factors which may influence the efficiency of transfer of TPP up the food web. Intuitively, we might expect that the most influential are likely to be

1 those at the lower trophic levels. Systematic variability in the complexity of the lower trophic level food web has the capacity to alter the effective trophic status of higher trophic levels and affect metabolic losses during the transfer of energy from primary producers to fish. Recognition of this is embodied in the concepts of “new” and “recycled” production (NP and RP), developed from the results of biogeochemical studies of mid-high latitude nitrogen limited open ocean ecosystems, particularly during the JGOFS programme.

New production in the ocean

The water column in the open ocean may be several km deep, but there is sufficient light for photosynthesis in only a thin layer (50-100m) at the surface (down to 0.1-1% of the sea surface irradiance). We refer to this as the photic zone. All of the dissolved inorganic nitrogen which is present as nitrate in the photic zone at the end of the winter is converted into living, and then dead particulate form as a result of photosynthesis, grazing and predation in the spring and summer. Dead particulate matter sinks out of the photic zone and disappears into the deep ocean. The depletion of nitrate in the photic zone sets up a vertical concentration gradient and nitrate diffuses from the underlying waters into the photic zone, supporting additional particulate matter production and export to the deep ocean, until declining irradiance in the autumn curtails photosynthesis. During the winter, vertical mixing recharges the photic zone with nitrate from the underlying waters. If the annual cycle of nitrogen in the photic zone is stationary, then the annual integrated vertical mixing and diffusion flux of nitrate into the photic zone from deep water must equal the annual particulate production of the photic zone food web and the sinking flux of detritus to the deep ocean. We refer to the primary production equivalent to the annual vertical mixing and diffusion flux as NP. However, NP does not equal TPP, because in the process of metabolizing the algal biomass equivalent to NP, the grazer and predator community in the photic zone excretes ammonia, which is then available to support additional primary production. We refer to the component of TPP supported by ammonia excreted by the heterotroph community as RP, and TPP = NP + RP.

Remineralisation and nitrification of the particulate matter sinking out of the photic zone occurs deep in the ocean and contributes to the pool of deep ocean nitrate. But, since this pool is large compared to the annual export flux from the photic zone, the effective recycling rate of nutrient back to the surface waters is extremely slow (on the order of hundreds of years) and the space scales correspondingly large. Thus there is a clear disconnection between the very long term cycling of nutrient between the deep ocean and the photic zone, and short term dynamics of nutrient uptake by autotrophs and excretion of ammonia by heterotrophs.

The ability to directly measure uptake rates of nitrate and ammonia using the stable isotope 15N incubation techniques provided a means of evaluating the relative importance of new and recycled production, and gave rise to the term “f-ratio”, defined as new/total production and estimated experimentally from

f = NP/TPP = NO3 uptake/((NO3 + NH4) uptake)

For much of the open ocean

0.1 < f < 0.5

One reason for the wide-spread use of the f-ratio in relation to open ocean biogeochemistry was the general conclusion from observations that the f-ratio increase monotonically with the rate of TPP – the greater TPP the greater the fraction that is new production. This provided the basis for “global” relations between the f-ratio, temperature and TPP (Laws et al, 2000) which have been used to create global inventories of carbon flux to the deep ocean.

In this paper we focus on the idea that the concept of NP could provide a closer relationship with integrated food web production and fisheries yield, than TPP. However, should we expect the generalizations and relationships established from ocean studies to hold for shelf regions? Are the concepts of NP and the f-ratio useful in such environments?

Measures of Production in shelf seas

The main fisheries interest in rates of primary production is in shelf seas rather than the open ocean. Hence, the concept of new versus total production has also been applied to shelf ecosystems (Richardson et al, 1998; Bisagni, 2003; Heath and Beare, 2008). However, there are a number of key differences.

In shallow coastal waters the photic zone may extend to the seabed and hence there is no separation in either space or time of the processes of nitrate uptake, and the mineralization and nitrification of organic nitrogen back to nitrate, such as exists in the ocean. Nitrate production, together with inputs from rivers, atmospheric deposition, anthropogenic discharges and horizontal mixing occur throughout the year in the photic zone. In addition, the process of denitrification which involves the microbial utilization of nitrate as a source of oxygen in anaerobic environments and the release of gaseous nitrogen to the atmosphere, is a potentally significant loss term in shallow waters. In estuarine systems in particular, denitrification is a very significant term in the nitrogen budget. In the ocean, denitrification must be confined to deep waters where it cannot compete with phytoplankton for the available nitrate. The dynamics of nitrate concentration in the coastal water column will therefore be complicated, and reflect the balance between external inputs, competition between phytoplankton and nitrifying bacteria for the ammonia produced by remineralisation of dead organic matter, and between phytoplankton and denitrifying bacteria for nitrate. We could consider the annual influx of external dissolved inorganic nitrogen plus the annual mineralisation flux from dead matter to ammonia as being equivalent to NP in the ocean. But, what then is the shallow shelf equivalent of RP? Ammonia excreted by herbivores and predators in the food web (the metabolic consequence of herbivorous grazing and predation) is indistinguishable from the mineralization products of dead organic material. Hence, some

3 of the autotrophic nitrate uptake in shallow coastal waters may be equivalent to RP in the ocean model, and some of the ammonia uptake may be equivalent to NP.

As one moves out from shallow coastal waters into deeper shelf areas, the space-time separation between the nitrogen uptake and recycling systems should begin to emerge, as the photic zone thickness becomes a smaller fraction of the total water column depth, and the near-seabed layers become more isolated from the surface. Nevertheless, its is clear that the simple definitions of new production and the f-ratio developed from open ocean studies, although potentially useful on the shelf, do not obviously provide a simple explanation of yields by shelf sea higher trophic levels. It is necessary to consider the structure of the microbial food web as a function of the physical topography of the sea bed in any particular region.

We use the schematic in Fig. 2 to illustrate the complexity of nitrogen dynamics on high latitude shelves such as Georges Bank or the North Sea. The three vertical mixing regimes –Stratified, Transition and well Mixed – are defined by frontal processes, specifically the development of tidal mixing fronts (Simpson and Hunter, 1974: Pingree et al, 1978; Bisagni, 2003). We can reasonably consider that the annual autotrophic uptake of dissolved inorganic nitrogen (DIN) is confined to the spring and summer months, so that remineralisation essentially resets dissolved nutrient loads during the winter. If we further assume that a) most of the labile organic matter in the system from the previous year is remineralised before the onset of the spring bloom, and b) that all of the autotrophic production is grazed by herbivores, then depth integrated mass of DIN at the onset of the bloom, effectively sets a limit to the net amount of material that can be transferred up the food web – the equivalent of NP in the ocean situation. Added to this should be the mass of DIN added to the water column during the spring and summer autotrophic production period. TP is then this NP equivalent, plus the cumulative ammonia excretion by the food web during the spring and summer. As we move out into deeper shelf waters, the proportion of pre-bloom depth integrated DIN which may be accessible to phytoplankton in the photic zone during the year becomes more dependent on vertical diffusion and mixing and hence less easy to determine practically.

We can define five possible measures in shelf waters which might be equivalent to NP in the ocean, and hence directly relatable to integrated food web production and potential fishery yield The conventional unit for biological production is carbon, but this can be inferred from nitrogen units using the Redfield ratio (C/N = 6.625).. (1) NIP. The experimental measure of nitrate uptake from time series of incubation experiments, as described for open ocean studies, (2) PNP. Potential New Production, determined as the sum of the vertical flux of nitrate into the photic layer and the decrease of nitrate in that layer. This is calculated from vertical profiles of nitrate and estimates of vertical mixing (Bisagni, 2003) (3) MMP. Defined as the difference, on an annual basis, between the maximum and minimum values of nitrate, integrated from surface to seabed (Heath and Beare, 2008).

4 (4) IP. Defined as the rate of input of inorganic N (nitrite + nitrate + ammonia) into the region from rivers, atmospheric deposition and exchange across the shelf edge. Calculated from river and deposition data, and estimates of advection across the shelf edge (Heath and Beare, 2008). (5) MMIP. Combination of MMP plus that part of IP occurring during spring and summer (Heath and Beare, 2008)

Within this framework we can define total (sometimes called net) production as, (6) TPP. The annually integrated sum of uptake rates of nitrate plus ammonia. Usually TP is estimated from satellite color data calibrated to carbon using 14C or 15N tracer uptake experiments.

It will be apparent from Fig. 2 that these five definitions for shelf sea new production are not equivalent. The open ocean assumptions of purely vertical processes and an effectively infinite deep reservoir do not hold. Also the lateral input of nutrients to shelf waters from across the shelf edge requires a corresponding flux of water out of the system, usually from the surface layer, that will carry with it nutrients and plankton. Further, each measure of new production is inapplicable to differing degrees and in different ways for the three mixing regimes in Fig. 2.The first three definitions may converge in large well stratified regions such as the Northern North Sea, but not on Georges Bank where flux out of the system is a major factor (Steele et al, 2007) for the different mixed regimes.

Comparison of the North Sea and Georges Bank

To illuminate the issues we compare estimates of new and total production for two North Atlantic shelf ecosystems, Table 2. We use PNP for Georges Bank (Bisagni, 2003) and MMIP for the North Sea (Heath and Beare, 2008). Both can be related to the physical geography of the systems in terms of the three mixing regimes in Fig. 2. In both systems the total production TPP increases with progression from the deep stratified regions to shallow well mixed areas. According to the open ocean paradigm, one would expect the f-ratio to increase correspondingly, with largest values in the well mixed region. This does not happen in either system, but there is no alternative generic pattern in the f-ratios. For Georges Bank, the largest ratio is in the Transition zone; in the North Sea it occurs in the Stratified Zone.

These differences can be explained, at least qualitatively, by the very different topography of the two systems. Georges Bank is a small, off-shore, bank with no river input to the well mixed zone, and the Transition zone is close to the shelf edge. The North Sea has significant river inputs to the well mixed zone and a Transition zone which is isolated from shelf-edge inputs by a wide Stratified zone. The result is that the Transition zone has maximum new production on Georges Bank and minimum in the North Sea.

Discussion

5 It is clear that our definition (1) of new production (NIP) is least likely to represent the shelf-sea equivalent of NP in the oceans, and hence the annual flux of nitrogen up the food web. In well mixed coastal waters a significant component of annual nitrate uptake must include within-year recycled nutrient due to the close spatial and temporal connection between uptake and nitrification processes. Moving out into stratified outer shelf waters, the separation of processes must widen, but never sufficient to make NIP an unambiguous measure of production on the shelf. In fact, there is increasing evidence that nitrification may not be completely negligible even in the open ocean photic zone (Ward, 2000: Yool et al, 2008), which means that experimentally derived f-ratios could generally under-estimate recycling.

Similarly, our definition (4, IP), which is superficially equivalent to the vertical mixing and diffusion flux in the ocean, cannot be wholly taken as the shelf equivalent of NP, since a significant component of the winter recharge of the photic zone is due to local remineralisation of organic matter produced during the previous summer. In the ocean, this process is spatially and temporally disconnected from the upper layers but is ultimately responsible for maintaining the vertical mixing and diffusion flux.

Our definitions 2, 3 and 5 (PNP, MMP and MMIP) must come closest to a shelf equivalent of NP. However, PNP assumes that there is no nitrification of ammonia to nitrate taking place in the photic zone and no external river, atmospheric or lateral inputs of inorganic nitrogen to the surface waters. MMP similarly assumes that there is no nitrification during the spring and summer months, and no external inputs. The issue of external inputs is resolved by MMIP, but the assumption of no nitrification during spring and summer remains an issue. Nevertheless, leaving aside the issue of within-year nitrification, we might expect the combined term PNP+IP, or MMIP in shelf waters to provide the closest equivalents to ocean NP.

It is clear from our comparison of Georges Bank and the North Sea that any relationships between TPP and the f-ratio which may exist in the open ocean, do not hold in the shelf seas. Nevertheless, we hypothesise that our shelf sea equivalent of the f-ratio (PNP/TPP around Georges Bank, and MMIP/TPP in the North Sea) is an expression of the efficiency of transfer of nutrient up the food web. We base this hypothesis on the idea that RP essentially reflects the metabolic requirements of the ecosystem since it is based on excreta and the within-season remineralisation of organic matter, whilst the concept of NP represents the potential for growth and reproduction of higher trophic levels (Horne et al., 1989). Hence, a low f-ratio indicates a high rate of metabolic turn-over in the system relative to primary uptake, and by implication a low transfer efficient up the food web.

If the take the tidally mixed, transition and summer stratified zones as defining physical structures of shelf seas, then our comparison of Georges Bank and the North Sea also shows that there is no generic relationship between these zones and the f-ratio or food web efficiency. More likely, the key factors are the distribution of external nutrient inputs between mixed, transition and stratified zones, and the extent to which the benthis system has direct access to microbial production taking places in the water column. In the 1960’s and’70’s, a major difference between the North Sea and Georges Bank lay in the ratio of

6 demersal to total fish production (Cohen et al, 1982); with the ratio for Georges Bank being approximately twice that for the North Sea. The Transition zone is the largest component on Georges Bank and has the highest new production. Because the mixing regime in the Transition zone extends to the bottom, the suspension feeding benthos are a major consumer of pelagic microbial production, accounting for the large production of benthivorous fish (Steele et al, 2007). Similarly, in the Celtic Sea, a relatively high proportion of the region is occupied by the Transition zone (Pingree et al, 1978) and this may may explain the observation (Heath, 2005a, b) that this region has relatively large benthic production compared with the North Sea.

More generally these comparisons demonstrate that detailed analysis of the microbial food web in the context of physical regimes is essential for a proper understanding of the composition of upper food webs and production by different fish guilds in relation to primary production. Failure to take these factors into account has probably let to the discrepancies in relationships between primary production and fish production that have arisen from earlier studies

References

Bisagni, J.J. (2003). Seasonal variability of nitrate supply and potential new production in the Gulf of Maine and Georges Bank regions. J. . Geophys. Res. 108, c11: 8015.

Chassot, E., Melin, F., Le Pape, O. and Gascuel, D. (2007). Bottom up control regulates fisheries production at the scale of eco-regions in European seas. Mar. Ecol. Prog. Ser. 343: 45-55.

Cohen, E.B., Grosslein, M.D. , Sissenwine, M.P. (1982). Energy budget of Georges Bank, Can. Spec. Publ. Fish. Aquat. Sci. 59: 95-107.

Dugdale, R.C. and Goering, J.J. (1967). Uptake of new and regenerated forms of nitrogen in primary productivity. Limnol. Oceanogr. 12: 196-206.

Eppley, R.W., B. J. Peterson (1979). Particulate organic matter flux and planktonic new production in the deep ocean. Nature 282:677-680.

Heath, M.R. (2005a). Changes in the structure and function of the North Sea fish food web, 1973- 2000, and the impacts of fishing and climate. ICES J. Mar. Sci. 62: 847-868.

Heath, M.R. (2005b). Regional variability in the trophic requirements of shelf sea fisheries in the Northeast Atlantic, 1973-2000. ICES J. Mar. Sci. 62: 1233-1244.

Heath, M.R. and Beare, D.J. (2008). New primary production in northwest European seas, 1960- 2003. Mar. Ecol. Prog. Ser. 363: 183-203.

Horne EPW, Loder JW, Harrison WG, Mohn R, Lewis MR, Irwin B, Platt T (1989). Nitrate supply and demand at the Georges Bank tidal front. Scient Mar. 53: 145-158

7 Iverson, R.L. (1990). Control of marine fish production. Limnol. Oceanogr. 35:1593- 1604

Laws, E.A., Falkowski, P.G., Smith,W.O, Ducklow,H., McCarthy, J.J. (2000). Temperature effects on export production in the open ocean. Global Biogeochem. Cycles 14:1231-1246.

Nixon, S.W. (1988). Physical energy inputs and the comparative ecology of lake and marine ecosystems. Limnol. Oceanogr. 33:1005-1025.

Pingree, R.D., Holligan, P.M., Mardell, G.T. (1978). The effects of vertical stability on phytoplankton distributions in the summer on the north-west European shelf. Deep-sea Res. 25, 1011-1028

Richardson, K, and Pederson, F.B. (1998). Estimates of new production in the North Sea: consequences for temporal and spatial variability of phytoplankton. ICES J. Mar. Sci. 55:574-580.

Simpson, H.J., Hunter, J.R. (1974). Fronts in the Irish Sea. Nature, 1250, 404-406.

Steele, J. H. and Collie, J. S. (2005). Functional diversity and stability of coastal ecosystems. In The Sea, Vol. 13 (eds. Robinson, A.R., Brink, K.) Harvard U. P., Cambridge, Mass. 783-817.

Steele, J., Collie, J., Bisagni,J., Fogarty, M., Gifford, D., Link, J., Sieracki, M., Sullivan, B., Beet, A., Mountain, D., Durbin, E., Palka, D., and Stockhausen, W. (2007). Balancing end-to-end budgets of the Georges Bank ecosystem. Prog. Oceanogr. 74, 423-448.

Ward, B. B. (2000). Nitrification and the marine nitrogen cycle, p. 427-454. In D. L. Kirchman (ed.), Microbial ecology of the oceans. Wiley-Liss, New York, N.Y.

Yool, A., Martin A.P., Fernandez, C. and Clark, D.R. (2008). The significance of nitrification for oceanic new production. Nature, 447: 999-1002.

Table 1. Fish yield, F (g wet wt. m-2 .yr-1 ) as a function of Net (or total) primary production, TPP (g C. m-2 .yr-1 ) from three studies

F = - 0.8 + 0.024 * TPP Nixon (1988) F = - 3.7 + 0.095 * TPP Iverson (1990) F = - 1.5 + 0.009 * TPP Chassot (2007)

Table 2. Annual New (PNP, MMIP) and total (TPP) production, and resultant f-ratios For Georges Bank and the North Sea.

Prodn Stratified Transition Mixed

PNP 79 181 91 Georges Bank TPP 302 343 399 f-ratio 0.27 0.41 0.23

MMIP 57 39 56 North Sea TPP 110 160 220 f-ratio 0.52 0.24 0.26

Figure 1. Schematic representation of a coastal ecosystem in relation to time scales. PP= primary production. ML=microbial loop (from Steele and Collie, 2005).

10 Mixed Transition Stratified River Flux Input Out

50 m Deep Input

100 m

Figure 2. A schematic representation of the vertical mixing regimes on the continental shelf, the nutrient inputs and the output flux of organic matter (adapted from Steele et al, 2007).