Pergamon

Adv. Space Res. Vol. 18, No. 7, pp. (7)117-(7)128, 1996 Copyright © 1995 COSPAR

An existing ecological model of DMS production has been extended and applied to the spring-summer period in the Subantarctic Southern Ocean. The model predicts that production of phytoplankton and dissolved DMS will increase during spring to reach a maximum in summer consistent with the atmospheric data collected at the Cape Grim baseline station. Archival Coastal Zone Color Scanner satellite imagery has been used to define the seasonal range in phytoplankton concentration in the study region and validate model predictions. Local measured wind and sea temperature data have been used to calculate the DMS transfer velocity which is used to compute the sea-to-air flux of DMS. The seasonal trend in predicted DMS flux is in good agreement with the flux estimates made from observations.

Dimethylsulfide (DMS; CH3SCH3 ) is an important sulfur-containing trace gas produced by some classes of marine phytoplankton /1; 2; 3; 4 /. DMS is present in oceanic surface waters at concentrations sufficient to sustain a considerable net flux to the atmosphere, which is currently estimated to be 0.5 ˜ 0.3 Tmol S yr-1 compared with the global natural (marine + terrestrial + volcanic) flux estimate of sulfur to the atmosphere of 0.78 Tmol S yr-1 /5/.

Charlson et al. /6/ have suggested that a major source of cloud condensation nuclei (CCN) over the oceans is the DMS produced by planktonic algae in seawater. DMS is oxidized in the atmosphere to form non-sea-salt sulfate ( nss SO42- ) and methanesulfonate (MSA) aerosols. Because the formation of clouds is sensitive to CCN density, it has been postulated that biological regulation of the climate is possible by affecting albedo and thus the Earth's radiation budget. This would be due to the effect of temperature and solar radiation on phytoplankton growth; however, it is still unclear whether the change in albedo will cause a positive or negative feedback on climate.

The Southern Ocean is relatively unpolluted and thus the production of sulfate aerosols will be mainly due to the biogenic source of DMS. Measurements made at Cape Grim, Tasmania (400 41'S, 1440 41'E) by Ayers et al. /7/ over a twenty month period have confirmed the connection between atmospheric DMS and aerosol sulfur species - a significant part of the Charlson et al. hypothesis. Sampling at Cape Grim was carried out at 94 m above sea level, only during 'baseline' conditions - periods during which the 10m wind direction is from the southwest quadrant (1900 - 2800 ). A pronounced seasonal cycle in atmospheric DMS was detected with an amplitude of about 10-20. A similar cycle in MSA and n.s.s sulfate was also evident suggesting a tight coupling between DMS, MSA and n.s.s. sulfate aerosols.

While there are very limited measurements of aqueous DMS in the ocean southwest of Cape Grim, northern hemisphere data on oceanic DMS concentrations / 8; 2; 9; 10/ show that average surface seawater DMS concentration may display a seasonal variation of up to a factor of 50 in mid and high latitudes, from a typical winter value of 0.2 nM to a summer maximum of 10 nM. Measurements made by McTaggart and Burton /11/ in the Southern Ocean among 40-530S during the 1988-89 austral summer recorded mean January DMS of 20.5 nM, a very high value compared to the global mean concentration of about 3nM /12/ or the value of 6.8 nM found in the North Sea during the summer of 1985 /2/.

Here we present a modelling study of the oceanic production of DMS and its flux to the atmosphere under conditions typical of the Subantarctic Southern Ocean southwest of Cape Grim, with the aim of understanding the factors responsible for the amplitude of the measured seasonal cycle in atmospheric DMS. The model of DMS production developed by Gabric et al. /13/, henceforth referred to as the GMSK model, has been extended to include light and temperature effects on phytoplankton growth and the sea-air flux has been formulated in terms of wind speed and sea surface temperature.

The structure of the GMSK model is given in Figure 1 with arrows representing fluxes of nitrogen and sulfur through the food web. The biotic compartments comprise a generic autotroph (phytoplankton), planktonic bacteria which metabolise DMS and its precursor dimethylsulfonium propionate (DMSP), and three heterotrophs. For simplicity, no higher trophic levels are considered although zooplankton export through grazing by fish is included. The model state variables are vertically averaged over the mixed layer depth. The model's ecological structure reflects the current thinking on the role of microorganisms in elemental recycling /14;15 / and the important part played by aerobic bacteria in DMS turnover in the water column /16 /.

In the extension of the model described here, phytoplankton growth rate is assumed to be affected by light, temperature and nutrient availability. Diurnal variability in the light field is also included. Loss of particulate organic matter (both N and S) occurs by sedimentation below the thermocline. Nitrogen is regenerated via the microbial loop, however to limit the number of rate parameters in the model we have not distinguished between nitrate and ammonium in the nutrient uptake formulation.

Loss of dissolved DMS occurs by photo-oxidation /17/ and ventilation to the atmosphere. Particulate DMS/DMSP can also be lost by sinking of fecal material or dead cells below the mixed layer. Biomass units are all mg N m-2, with the sulfur species measured in mg S m-2. The specific form of the fluxes and parameter values is given in the Appendix with further details available in /13/.

Study Window

The region of the Southern Ocean in which DMS production will affect the measurements of atmospheric DMS made at Cape Grim depends on wind strength, and the lifetime of gaseous DMS in the marine boundary layer. Data are collected only when the wind direction is from 1800-2900, thus avoiding potentially polluted continental air masses.

Wind speeds vary from a monthly mean value of 41 km h-1 in January to 26 km h-1 in June, so that an air parcel may travel up to 1000 km per day. Since the atmospheric lifetime of DMS varies from approximately half a day in summer to about a week in winter, the monthly mean DMS concentrations observed at Cape Grim may reflect oceanic production in a quadrant well over 1000 km west and south of the sampling station.

Calculation of the Transfer Velocity

In order to adapt the GMSK model to conditions pertaining in the Southern Ocean, the air-sea exchange coefficient or transfer velocity (a constant the original version of the GMSK model) has been parameterised in terms of wind speed and ocean temperature. The wind speed (w ) dependence of the transfer velocity kw (cm h-1) was calculated from the equations given by Liss and Merlivat /18/ which were derived for CO2 at 200C and a Schmidt number (Sc) of 600,

kw = 0.17w		for w <= 3.6		(1a)
kw = 2.85w - 9.65		for 3.6 < w <= 13		(1b)
kw = 5.9w - 49.3		for w > 13		(1c)

These equations must be rescaled to apply for DMS and different Sc. Liss and Merlivat /18/ assume that kw is proportional to Sc-2/3 , for wind speeds less than 3.6 ms-1 , and Sc-1/2 for higher wind speeds. Rescaling equations (1) and enforcing piecewise continuity at the wind speed boundaries gives,

kw = a 0.17w		for w <= 3.6		(2a)
kw = b ( 2.85w - 10.26 ) + 0.612 a		for 3.6 < w <= 13		(2b)
kw = b ( 5.90w - 49.91 ) + 0.612 a		for w > 13		(2c)

with a = (600/Sc)2/3 and b = (600/Sc)1/2.

For a given gas, Sc varies with water temperature, decreasing as the temperature increases. The dependence of Sc on sea surface temperature (SST) for DMS was presented graphically by Erickson et al. [1990] and a cubic polynomial has been fitted to that data for use in the model,

Sc = 3628.5 - 234.58(SST) + 7.8601 (SST)² - 0.1148 (SST)³

(3)

Sea surface temperature in the Southern Ocean southwest of Cape Grim has been modelled as a periodic function with a period of one year,

SST = 0.5{( SSTmax + SSTmin ) + (SSTmax - SSTmin) cos [(t - tm) PI/180)]}

(4)

where SST_max and SST_min are the maximum and minimum annual temperatures, t the time in Julian days and tm the day on which the maximum temperature occurs. From air temperature data recorded at Cape Grim, the maximum and minimum sea surface temperatures are estimated to be 14.6°C and 9.4°C, respectively, with the maximum occurring on day 17. Thus from (3), Sc varies from a value of 1522 (at 14.6°C) to 2022 (at 9.4°C).

Photosynthesis

The original GMSK model formulation assumed that phytoplankton growth was not limited by either light or temperature, an appropriate simplification for predictions on the time scale of a bloom episode but not valid when seasonal variability is to be investigated. Assuming that both availability of light and the ambient temperature can effect the specific nutrient uptake rate, m, at any given time - the multiplicative model /19/,

m = VN RL RT

(5)

where V_N (h^-1) is the nitrogen-specific nutrient uptake rate following the Michaelis-Menten form given in the original GMSK formulation, and R_L and R_T are dimensionless light and temperature limitation coefficients, respectively.

Field measurements suggest nitrate levels in the Subantarctic Southern Ocean are typically very high in late winter 11.5 mM and drop to 2.1 mM by the end of the austral summer. The nitrate half-saturation concentration Ks varies with algal species size and ambient nutrient status, however oceanic species tend to have values in the range 0.1 to 1.0 mM with higher values (>4 mM) in coastal and eutrophic waters /20/. A Ks value of 1.0mM was used in the GMSK model and in the absence of better taxonomic data has been left unchanged.

A radiation model /21/ has been calibrated with meteorological data on cloudiness and surface radiation collected at Cape Grim and used to compute the incident solar irradiance at the sea surface Is, in the range 350-700 nm (photosynthetically active radiation - PAR). The ratio of PAR to total incoming solar radiation has been estimated to be in the range 0.42 to 0.48 for mid to high latitudes /22/.

In the Subantarctic ocean the mixed layer depth varies seasonally from about 100 m in summer to over 400 m in winter. The euphotic zone depth is fairly constant at around 90m for the whole year. For simplicity the euphotic zone averaged irradiance Ie has been used, assuming that light is exponentially attenuated with depth, and that the irradiance at the bottom of euphotic zone (z = Ze) is 1% of the surface value, thus the depth-integrated irradiance is

	        Ze					
Ie = (Is / Ze ) | e -kz  dz  =  0.21 Is                    (6)
                0

Light limitation of phytoplankton growth may be modelled in a number of ways. For example, Smith /23/ has suggested,

RL = P / Pmax = (I / Ik )[1 + (I/Ik)2 ] -0.5

(7a)

where P is the gross photosynthesis rate, Pmax the maximum photosynthesis rate, and I_k the saturating irradiance /24/. Steele /25/ used a formulation which includes photoinhibition at high light intensities,

RL = (I / Iopt) exp{ 1 - I/Iopt }

(7b)

where I_opt is the light intensity at which Pmax occurs.

There is a clear temperature effect on the growth rate of phytoplankton with the maximum rate of growth roughly doubling every 100C increase in sea temperature /26/. The limitation due to temperature is given by,

RT = (mo/mmax) exp(0.063T)

(8)

where m_o is 0.84 day^-1 and T is the temperature in degrees Celsius. For the maximum sea temperature in the study window (14.6°C), the maximal growth rate mmax is 2.1 day^-1 . By comparison , for winter sea temperature of 9.4°C, the growth limitation RT is 0.72.

Remote Sensing Data

Chlorophyll pigment concentration for the Southern Ocean southwest of Cape Grim was derived from archival Coastal Zone Color Scanner (CZCS) data. The CZCS 1978-1986 monthly composite archive was processed for the 1000 x 1000 km study window 41 - 51°S and 133.3 - 143.2°E and the results for spatially averaged chlorophyll-like pigment concentration are shown in Figure 2a and 2b. While there is no well-defined peak in the seasonal cycle, the minimum value in the northern section occurs over August to September while in the southern half the minimum occurs in June. In both sections variance is high throughout the year except during August and September. The monthly mean pigment concentration is in the range 0.2 - 0.3 mg chl m^-3 during the December to March period.

While there is little data on the seasonal cycles of phytoplankton species in the Subantarctic Southern Ocean, it appears that diatoms are an important component of the total flora, especially further south. McTaggart and Burton /11/ report that coccolithophores form a significant portion of the total phytoplankton community north of the Antarctic Convergence (53°S).

Fig. 2. Monthly average pigment concentration as derived
from CZCS 1978-86 composited data (a) 410S - 460S (b) 460S - 510S

Model Predictions

Wind speed records collected at Cape Grim were statistically analysed and the monthly mean speed E(w) and standard deviation s are given in Table 1.

The wind data were used to compute the variation in transfer velocity over a year and the results plotted in Figure 3. The transfer velocity varies from a maximum value of 3.4 m d^-1 in January to a minimum in July of 1.6 m d^-1. From equations (2) it is clear that kw is directly proportional to wind speed and inversely proportional to Schmidt number, which is a minimum in January. Wind speed is a maximum in January and the combined effects of high wind and warm sea temperatures result in a maximum in k_w during summer.

Erickson et al. /28/ computed global DMS transfer velocity based on global climate model generated wind and temperature fields. Interestingly, their predictions for the Southern Ocean near Cape Grim are higher in winter ( 6 m d^-1 ) than in summer (3.6 m d^-1), although their summer prediction is very close to that presented here. Given that our findings were based on locally derived data on wind and air temperature, the winter DMS flux given by Erickson et al. may be seriously overestimated.

Fig. 3 Predicted variation in transfer velocity based on wind speed data collected at Cape Grim

The extended GMSK model was integrated forward in time from early October (Julian day 275) through to mid-May (Julian day 500). We have assumed that the oceanic mixed layer during spring/summer is 100 m deep. The biotic state variables were all initialised to arbitrarily low values and the sulfur species (DMSP and DMS) were initialised to zero. For the high nitrate values typical of the Southern Ocean in early spring nitrogen is unlikely to limit phytoplankton growth so that phytoplankton growth rate is largely controlled by light and temperature.

Model predictions are known to be very sensitive to the algal cell DMSP content. Assuming that the phytoplankton community is dominated by diatoms and coccolithophores, we choose a S(DMSP):N cell ratio of 0.3 /13/. The depth-dependent model parameters were re-scaled to the 100m MLD and are given in the Appendix.

Several model runs were carried out and typical results for the growth of phytoplankton and the production of dissolved DMS are shown in Figure 4 a,b.

Fig. 4 Predicted variation in (a) phytoplankton biomass and (b) dissolved DMS

Gabric et al. /13/ have noted that the model predictions are very sensitive to the specification of the phytoplankton maximum nutrient uptake rate (k23) with both the timing and magnitude of DMS peaks being affected. This parameter will be species dependent and given the large area of ocean that can affect DMS measurements made at Cape Grim and the likely presence of a number of species in different stages of growth, the model predictions are clearly only an indicator of the actual cycle in DMS.

The phytoplankton growth follows a cyclical pattern with period of about 40 days and a gradual decline in the peak during autumn. DMS peaks ensue the algal peaks with a lag of about 1-2 days and follow a similar decline. DMS levels remain high for at least 10 days after phytoplankton have been completely grazed. The predicted mixed layer peak phytoplankton concentrations vary from almost 5000 to 800 mg Nm^-2 .

In order to compare the model predictions with the pigment concentrations detected in the CZCS imagery, a currency conversion from the model units of nitrogen to chlorophyll is required. Carbon:chlorophyll ratios can vary during the bloom cycle and also across algal taxonomic group. /20/.Assuming a C:Chl ratio of 50 and a C:N of 6, the model predictions correspond to chlorophyll concentrations in the range 6 to 1 mg m^-3. Mixed layer peak DMS concentrations vary from 19 to 4 mg S(DMS)m^-2 which correspond to the range 6 to 1.2 nM. This is similar to the range measured by Bates et al. /8/ in the North Pacific and in the Baltic /9/.

Model predictions for the sea-to-air flux of DMS are shown in Figure 5 together with independent estimates of the monthly average flux made by Ayers et al. /29/ using the Cape Grim atmospheric measurements and data on the rate of gaseous DMS removal through reaction with the hydroxyl radical. Given the limitation that our model assumes a biologically homogeneous ocean, the correspondence between the two sets of predictions is encouraging. The flux increases toward a maximum at the end of the year and then decreases in autumn which is identical to the trend in atmospheric DMS at Cape Grim.

Fig. 5 Model predictions for DMS flux

It should be noted that the model predictions are based on the assumption of a spatially uniform study region. Clearly this is a gross simplification given the extent of ocean which will affect DMS measurements at Cape Grim. The phytoplankton population over such a large area will likely be heterogeneous, implying different cell DMSP content, and in different stages of the growth cycle. Thus, while the flux predictions display a similar periodicity to the dissolved DMS concentration, the effect of heterogeneity in the phytoplankton assemblage and asynchronous blooming would be to fill in the troughs and give a more uniform flux over time as is indeed found in the experimental record.

1. M.D. Keller, WK Bellows and RL Guillard. Dimethyl Sulfide production in marine phytoplankton. In, Biogenic sulfur in the environment, Saltzman ES and Cooper WJ (eds), American Chemical Society, Washington, DC , 1989.

2. S.M. Turner, G Malin , PS Liss , DS Harbour and PM Holligan. The seasonal variation of dimethylsulfide and dimethylsulfoniopropionate concentrations in nearshore waters. Limnol. Oceanogr., 33(3), 364-375, 1988.

3. RL Iverson, FL Nearhoof and MO Andreae. Production of dimethylsulfonium and dimethylsulfide by phytoplankton in estuarine and coastal waters. Limnol. Oceanogr., 34(1), 53-67,1983.

4. S. Belviso, S.-K. Kim, F. Rassoulzadegan, B.Krjka, B.C.Nguyen, N.Mihalopoulos, and P.Buat-Menard. Production of dimethylsulfonium propionate (DMSP) and dimethylsulfide (DMS) by a microbial food web. Limnol. Oceanogr., 35(8): 1810-1821, 1990.

5. T.S. Bates, B.K. Lamb, A. Guenther, J. Dignon, and R.E. Stoiber. Sulfur emissions to the atmosphere from natural sources, J. Atmos. Chem., 14, 315-337, 1992.

6. R.J. Charlson, J.E. Lovelock , M.O. Andreae and S.G. Warren. Oceanic phytoplankton, atmospheric sulphur, cloud albedo and climate. Nature, 326, 655-661, (1987).

7. G.P. Ayers, JP Ivey and RW Gillett. Coherence between seasonal cycles of dimethylsulphide, methanesulphonate and sulphate in marine air. Nature, 349, 404-406, 1991.

8. T.S. Bates, JD Cline , RH Gammon and SR Kelly-Hansen. Regional and seasonal variations in the flux of oceanic dimethylsulfide to the atmosphere. J Geophys. Res., 92, C3, 2930-2938, 1987.

9. C. Leck, U Larsson , LE Bagander , S Johansson and S Hajdu. DMS in the Baltic Sea - Annual variability in relation to biological activity. J Geophys. Res., 95 (C3), 3353-3363, 1990.

10. B.C. Nguyen , N Mihalopoulos and S Belviso. Seasonal variation of atmospheric dimethylsulphide at Amsterdam Island in the Southern Indian Ocean, J. Atmos Chem., 11, 123-141, 1990.

11. A.R. McTaggart, and H Burton. Dimethyl sulfide concentrations in the surface waters of the Australasian Antarctic and Subantarctic oceans during an austral summer. J. Geophys. Res., 97, 14407-14412, 1992.

12. M.O. Andreae, and W.R. Barnard. The marine chemistry of dimethylsulphide. Mar. Chem., 14: 267-279, 1984.

13. A.J. Gabric, Murray N, Stone L and Kohl M. Modelling the production of dimethylsulfide during a phytoplankton bloom. J. Geophys. Res., 98, C12, 22805-22816, 1993.

14. F. Azam, T Fenchel, JS Gray, LA Meyer-Reil and F Thingstad. The ecological role of water-column microbes in the sea., Mar. Ecol. Prog. Ser., 10, 257-263, 1983.

15. T. Fenchel, Marine plankton food chains. Ann. Rev. Ecol. Syst., 19, 19-38, 1988.

16. R.P. Kiene and TS Bates. Biological removal of dimethylsulfide from seawater. Nature, 345, 702-705, 1990.

17. P. Brimblecombe and Shooter D. Photo-oxidation of dimethylsulfide in aqueous solution. Mar. Chem.., 19, 343-353, 1986.

18. P.S. Liss and L Merlivat. Air-sea gas exchange rates: Introduction and synthesis. In, The role of air-sea exchange in geochemical cycling, P. Buat-Menard (ed) 113-127, Reidel, Hingham, MA,1986.

19. T. Platt, Denman,K.L. and Jassby, A.D. Modeling the productivity of phytoplankton, in The Sea, Vol.6, E.D.Goldberg (ed.) New York, Wiley, 807-856, 1977.

20. T.R. Parsons , Takahashi M and Hargrave B. Biological Oceanographic Processes. 3rd Ed, Pergamon Press, 1984.

21. T.D.Brock, Caculating solar radiation for ecological studies, Ecol Model., 14, 1-19, 1981.

22. K.S.Baker and Frouin R. Relation between photosynthetically available radiation and total insolation at the ocean surface under clear skies. Limnol. Oceanogr. , 32(6), 1370-1377, 1987.

23. E.L. Smith. Photosynthesis in relation to light and carbon dioxide, Proc. Nat. Acad. Sci. Wash., 22, 504-511, 1936.

24. J.F. Talling. Photosynthetic characteristics of some freshwater plankton diatoms in relation to underwater radiation, New Phytol., 56, 29-50, 1957.

25. J.H. Steele. Environmental control of photosynthesis in the sea. Limnol. Oceanogr., 7, 137-150, 1962.

26. R.W. Eppley. Temperature and phytoplankton growth in the sea, Fish. Bull., 70, 1063-1085, 1972.

27. G. Jacques , Descolas-Gros C, Grall JR amd Sournia A. Distribution du phytoplancton dans le partie antarctique de l'Ocean Indienne en fin d'ete, Int. Rev. Res. Hydrobiol., 64, 609-628, 1979.

28. D.J. Erickson, Ghan SJ and Penner JE. Global ocean-to-atmosphere dimethyl sulfide flux., J. Geophys. Res., 95, D6, 7543-7552, 1990.

29. G.P. Ayers, Ivey JP, Bentley ST and Forgan BW. Dimethylsulphide in marine air at Cape Grim, 41_S, Tellus (in press)

APPENDIX

Definition of Model State Variables and Fluxes

X1	phytoplankton , mg Nm-2
X2	bacteria, mg Nm-2
X3	zooflagellates, mg Nm-2
X4	large protozoa, mg Nm-2
X5	micro and mesozooplankton, mg Nm-2
X6	dissolved inorganic nitrogen, mg Nm-2
X7	dissolved DMSP, mg S(DMSP) m-2
X8	dissolved DMS, mg S(DMS) m-2
F12	bacterial decomposition of phytoplankton, mg Nm-2d-1
F14	ingestion of phytoplankton by large protozoa, mg Nm-2d-1
F15	ingestion of phytoplankt. by micro-mesozooplankton,mg Nm-2d-1
F17	excretion of DMSP by phytoplankton, mgS(DMSP) m-2d-1
F18	excretion/metabolism of DMS by phytoplankt, mgS(DMS)m2d-1
F1W	sedimentation of phytoplankt below euphotic zone, mg Nm-2d-1
F23	ingestion of bacteria by zooflagellates, mg Nm-2d-1
F26	excretion of DIN by bacteria, mg Nm-2d-1
F34	ingestion of zooflagellates by large protozoa, mg Nm-2d-1
F36	excretion of DIN by zooflagellates, mg Nm-2d-1
F45	ingestion of protozoa by zooplankton, mg Nm-2 d-1
F46	excretion of DIN by large protozoa, mg Nm-2d-1
F47	excretion of DMSP by large protozoa, mg S(DMSP) m-2d-1
F56	excretion of DIN by micro and mesozooplankton, mg Nm-2d-1
F57	excretion of DMSP by zooplankton, mg S(DMSP) m-2d-1
F5W	sedimentation and export of zooplankton, mg Nm-2d-1
F61	DIN uptake by phytoplankton, mg Nm-2d-1
F62	DIN uptake by bacteria, mg Nm-2d-1
F72	DMSP biodegradation and uptake by bacteria, mg S(DMSP) m-2d-1
F78	Conversion of DMSP to DMS in water column, mgS(DMSP)m-2d-1
F82	DMS uptake/consumption by bacteria, mg S(DMS) m-2d-1
F8W	DMS photo-oxidation, adsorption/sedimentation, mgS(DMS)m-2d-1
F8A	DMS ventilation to atmosphere, mg S(DMS) m-2d-1

*F_ij is the flow between compartments X_i and X_j (A is atmosphere; W is water column)


Model Equations

dX1/dt	=	F61 - F12 - F14 - F15 - F1W
dX2/dt	=	F12 + F62 - F23 - F26 + a1 F82 + a2 F72
dX3/dt	=	F23 - F36 - F34
dX4/d t	=	F34 + F14 - F45 - F46
dX5/dt	=	F45 + F15 - F56 - F5W
dX6/dt	=	F26 + F36 + F46 + F56 - F61 - F62
dX7/dt/dt	=	F17 + F47 + F57 - F78 - F72
dX8/dt	=	F18 + b F78 - F 82 - F8W - F8A
With
F12	=	k1 X2 [ 1 - exp(-k2 X 1) ]
F14	=	k3 X1 X4
F15	=	k4 X1 X5
F17	=	k5 X1 g
F18	=	b g k6 X1
F1W	=	k7 X1
F23	=	k8 X3 [ 1 - exp(-k9 X2) ]
F26	=	k10 X2 + k11 [ F62 + F12 ]
F34	=	k12 X3 X4
F36	=	k13 X3 + k14 F23
F45	=	k15 X4 X5
F46	=	k16 X4 + k17 [ F34 + F14 ]
F47	=	k18 X4 g
F56	=	k19 X5 + k20 [ F15 + F45 ]
F57	=	k21 X5 g
F5W	=	k22 F56 + k32 X5
F61	=	k23 X1 [ 1 - exp(-k24 X6) ]
F62	=	k25 X2 [ 1 - exp(-k26 X6) ]
F72	=	k31 X7
F78	=	k27 X7
F82	=	k28 X8
F8W	=	k29 X8
F8A	=	k30 X8

Currency conversion factors a₁ and a₂ are given by 0.15 and 0.37, respectively, and g is the S(DMSP):N ratio for phytoplankton (species dependent ).



Model Parameters for a 100-m mixed layer

	Value	Pathway*	Units
k1	4.5	P-B*	day-1
k2	4.6e-4	P-B	m2 mg N-1
k3	2.6e-3	P-LP	m2 mgN-1d-1
k4	1.2e-3	P-Z	m2 mgN-1d-1
k5	0.01	P-DMSP	day-1
k6	0.0085	P-DMS	day-1
k7	0.15	P Sinking	day-1
k8	17	B-F	day-1
k9	1.38e-3	B-F	m2 mg N-1
k10	0.07	B-N	day-1
k11	0.63	B-N	...
k12	0.0156	F-LP	m2 mgN-1d-1
k13	0.05	F-N	day-1
k14	0.65	F-N	...
k15	1.2e-3	LP-Z	m2 mgN-1d-1
k16	0.05	LP-N	day-1
k17	0.65	LP-N	...
k18	0.01	LP-DMSP	day-1
k19	0.05	Z-N	day-1
k20	0.40	Z-N	...
k21	0.01	Z-DMSP	day-1
k22	0.15	Z sinking	...
k23	0.9	N-P	day-1
k24	5e-4	N-P	m2 mg N-1
k25	0.9	N-B	day-1
k26	9.24e-3	N-B	m2 mg N-1
k27	0.5	DMSP-DMS	day-1
k28	0.95	DMS-B	day-1
k29	0.27	DMS photo-ox	day-1
k30	see text	DMS - atmos	day-1
k31	1.0	DMSP-B	day-1
k32	0.05	Z export	day-1

---

For furthur information, contact Nicholas Murray, EI-JRC ,nicholas.murray@jrc.it

---

[ ME Unit ]

---