Technical Description of the CSIRO SERM Ecological Model

CSIRO Simple Estuarine Response Model1

Technical Description of the Ecological Model

Mark Baird2 CSIRO Land and Water 1 November 2001

Introduction The CSIRO Simple Estuarine Response Model (SERM) has been developed to provide a modelling tool for the Australian continent's approximately 1000 estuaries. To capture the dynamics of ecological processes over a wide variety of Australian estuaries, the ecological model had been developed to be both simple, and to be composed of parameters whose values are not expected to vary significantly between estuaries. To work towards this goal, many of the key biological processes have been described mechanistically. Mechanistic process descriptions include attenuation of light through the water column and benthic biota, nutrient uptake and light capture by algae, and encounter rates of phytoplankton and zooplankton. Other descriptions, such as sediment chemistry, seagrass growth and the effect of higher trophic levels, where processes are more complex, have been modelled empirically. The ecological model described in this document is a modification of the CSIRO Port Phillip Bay Environmental Study ecological model. In particular, the model is implemented using the same box model approach and software as used in the PPB study, and contains a similar partitioning of water column, epibenthic and sediment tracers. Further details can be found in Murray and Parslow (1997). The most succinct description of SERM can be found on the website (www.marine.csiro.au/serm). This document fills in the details. It is composed of sections detailing the state variables (p 3), parameters (p 4), constants (p 7), ecological processes (p 8) that make up the model. Then follows a description of the ecological model equations (p 13), physical model equations (p 17), ecological process descriptions (how the ecological model equations were derived, p 20) and a list of references (p 32). First, well define some terminology and abbreviations (p 2). The original components of the model are submitted for publication:
Baird, M. E., S. J. Walker, B. B. Wallace, I. T. Webster and J. S. Parlsow (submitted). Towards a mechanistic model of estuarine eutrophication. Estuarine, Coastal and Shelf Science.
1

This work was partially funded by the generous support of the National Land and Water Resources Audit, a program of the National Heritage Trust, Australia. 2 Corresponding address: School of Mathematics, UNSW, Sydney 2052 Australia (mbaird@maths.unsw.edu.au)

Technical Description of the CSIRO SERM Ecological Model

Terminology and Abbreviations System - for our purposes, the 'system' is the ecology of an estuary. Model - a framework (in our case mathematical) used to aid our understanding of the system. State Variables - state variables describe the state of the system. Indicators on the SERM interface are a single or combination of state variables. The output of the model is the value of each of these state variables at each time point. The value of a state variable is changed either by a process, or a forcing. An example is nutrient concentration, which is changed by the process of nutrient uptake, or the forcing of a point source load. Process - an underlying phenomena or collection of phenomenon. The major ecological processes are represented mathematically using the model equations. Processes change the value of state variables. For example, nutrient uptake changes nutrient concentration. Parameters - parameters are variables that are assumed to be constant within a particular system. There is some uncertainty as to the value parameters should take, and it is common to undertake a sensitivity analysis to investigate the effect of varying parameter values. An example of a parameter is the maximum growth rate of small phytoplankton. Constants - constants are variables which, to a high degree of accuracy, do not vary between systems or in time. Examples include gravitation acceleration of the earth and the molecular diffusivity of a chemical species. Forcings and Boundary Conditions - forcings are inputs into the model (i.e. they change the value of the state variables) that are dependent on processes external to the model system. These include sunlight, river inputs, and boundary conditions. Model Equations - model equations are mathematical representations of the processes linking state variables. They are made up of a combination of state variables, parameters and constants, and are formulated to represent, as well as is known (in some cases mechanistically, in other cases empirically) our best knowledge of what controls the rate of processes in the model. In particular, we are careful to ensure that the equations conserve mass. In other words, mass cannot be created or destroyed, but rather moves between the different state variables or across boundaries. Initial Conditions - each state variable must be given an initial condition, from which the simulation begins. Ideally, if the model is run to a stable annual oscillation, the final result should be independent of initial conditions. Mechanistic - a description of a process using an understanding of the underlying physical processes. Empirical - a description of a process determined using data from experiments or field observations.

Technical Description of the CSIRO SERM Ecological Model

State Variables The model contains 29 state variables given below. The name of the variable in the code, and its units are also given. Note that some of the state variables represent fixed quantities of carbon, nitrogen and phosphorus. Such state variables are quantified in units of nitrogen, but contain a fixed ratio of carbon to nitrogen to phosphorus (C:N:P) by atoms, given in brackets: State Variable Large phytoplankton (106:16:1) Small phytoplankton (106:16:1) Microphytobenthos (106:16:1) Seagrass (550:30:1) Macroalgae (550:30:1) Large zooplankton (106:16:1) Small zooplankton (106:16:1) Labile detritus at Redfield ratio (106:16:1) Labile detritus at Atkinson ratio (550:30:1) Refractory detrital carbon Refractory detrital nitrogen Refractory detrital phosphorus Dissolved organic carbon Dissolved organic nitrogen Dissolved organic phosphorus Ammonia Nitrate Dissolved inorganic carbon Dissolved inorganic phosphate Particulate inorganic phosphate (unfloc.) Particulate inorganic phosphate (floc.) Total Suspended Solids (unflocculated) Total Suspended Solids (flocculated) Dissolved oxygen Light Temperature Salinity Symbol PL PS MPB SG MA ZL ZS LDP LDB RDC RDN RDP DOC DON DOP NH NO DIC DIP PIP_unfloc PIP_floc TSS_unfloc TSS_floc DO I T S Units mg N m-3 mg N m-3 mg N m-3 mg N m-2 mg N m-2 mg N m-3 mg N m-3 mg N m-3 mg N m-3 mg C m-3 mg N m-3 mg P m-3 mg C m-3 mg N m-3 mg P m-3 mg N m-3 mg N m-3 mg C m-3 mg P m-3 mg P m-3 mg P m-3 kg m-3 kg m-3 mg O2 m-3 W m-2 C PSU

Technical Description of the CSIRO SERM Ecological Model

Parameters There are 66 biological parameters. The symbol in the code, a description, and the values used in the model are given below (parameter is dimensionless unless units are given). Parameter E_ZL E_ZS SG_KN SG_KP ML_PL_T15 ML_PS_T15 ML_MA_T15 ML_SG_T15 MQ_MPB_T15 MQ_ZL_T15 MQ_ZS_T15 FDG_ZL FDM_ZL FDG_ZS FDM_ZS F_LD_RD F_LD_DOM R_0_T15 R_D_T15 Dmax X_CHLN k_w k_DON k_DL k_TSS k_SWR_PAR Description Growth efficiency, large zooplankton Growth efficiency, small zooplankton Half-saturation of SG N uptake in the sediment Half-saturation of SG P uptake in the sediment Linear mortality rate, large phytoplankton (in sediment) Linear mortality rate, small phytoplankton (in sediment) Linear mortality rate, macroalgae Linear mortality rate, seagrass Quadratic mortality rate, microphytobenthos Quadratic mortality rate, large zooplankton Quadratic mortality rate, small zooplankton Fraction of large zooplankton growth inefficiency lost to detritus Fraction of large zooplankton mortality lost to detritus Fraction of small zooplankton growth inefficiency lost to detritus Fraction of small zooplankton mortality lost to detritus Fraction of labile detritus converted to refractory detritus Fraction of labile detritus converted to dissolved organic matter Sediment net respiration rate at which nitrification = 0 Sediment net respiration rate of denitrification maximum Maximum efficiency of the removal of N by nitrification followed by denitrification nitrogen to Chlorophyll a ratio in phytoplankton by weight Background light attenuation coefficient DON-specific light attenuation coefficient Detrital N specific light attenuation coefficient TSS specific light attenuation coefficient fraction of incident solar radiation that is PAR Value (with units) 0.38 0.38 5.0 mg N m-3 5.0 mg P m-3 0.14 d-1 0.14 d-1 0.00274 d-1 0.00274 d-1 0.0003 d-1 (mg N m-3)-1 0.02 d-1 (mg N m-3)-1 0.02 d-1 (mg N m-3)-1 0.25 0.25 0.25 0.25 0.19 0.01 200 mg N m-2 d-1 10 mg N m-2 d-1 0.7 7 mg N (mg Chl a)-1 0.1 m-1 0.0009 m-1 (mg N m-3)-1 0.0038 m-1 (mg N m-3)-1 30.0 m-1 (kg m-3)-1 0.43

Technical Description of the CSIRO SERM Ecological Model

Q10 PLumax PLrad PLabsorb PSumax PSrad PSabsorb MPBumax MPBrad MPBabsorb MAumax MAaA SGumax SGaA ZSumax ZSrad ZSswim ZLumax ZLrad ZLswim TKEeps cf Ub ks F_RD_DOM r_floc r_absP P_buff_sed P_abs_coef r_LDP_T15 r_LDB_T15 r_RD_T15 r_DOM_T15 Plank_resp Benth_resp

Temperature coefficient for rate parameters Maximum growth rate of PL at Tref Radius of the large phytoplankton cells Absorption coefficient of a PL cell Maximum growth rate of PS at Tref Radius of the small phytoplankton cells Absorption coefficient of a PS cell Maximum growth rate of MPB at Tref Radius of the microphytobenthos cells Absorption coefficient of a MPB cell Maximum growth rate of MA at Tref Nitrogen specific absorption cross-section of MA Maximum growth rate of SG at Tref Nitrogen specific absorption cross-section of SG Maximum growth rate of ZS at Tref Radius of the small zooplankton cells Swimming velocity for small zooplankton Maximum growth rate of ZL at Tref Radius of the large zooplankton cells Swimming velocity for large zooplankton Dissipation of turbulent kinetic energy in the water column drag coefficient of the benthic surface velocity at the top of the benthic boundary layer sand-grain roughness of the benthos fraction of refractory detritus that breaks down to DOM rate at which TSS flocculates above 10 PSU rate at which P reaches ab/desorbed equilibrium Phosphate buffering capacity of sediment Phosphate absorption coefficient Breakdown rate of labile detritus at 106:16:1 Breakdown rate of labile detritus at 550:30:1 Breakdown rate of refractory detritus Breakdown rate of dissolved organic matter Respiration as a fraction of max. growth rate Respiration as a fraction of max. growth rate

2.0 1.25 d-1 10 m 50000 m-1 1.25 d-1 2.5 m 50000 m-1 0.35 d-1 10 m 50000 m-1 0.2 d-1 1 x 10-3 m2 mg N-1 0.1 d-1 1 x 10-5 m2 mg N-1 3 d-1 12.5 m 100 m s-1 0.375 d-1 50 m 200 m s-1 1 x 10-6 m2 s-3 0.005 0.1 m s-1 0.1 m 0.05 0.01 d-1 1.0 d-1 2650.0 kg kg-1 2.0 m3 kg 0.1 d-1 0.1 d-1 0.0036 d-1 0.00176 d-1 0.025 0.025

Technical Description of the CSIRO SERM Ecological Model

Two further biological parameters had estuary-specific values: K_CDOM Tref Light attenuation of freshwater inflow Reference temperature for biological processes estuary dependent [m-1] estuary dependent [C]

In the SERM model, the estuary-specific biological parameters are set by choosing an estuarine parameter value. In addition, there are 14 parameters within the physics model which relate to biological components of the model: w_PL w_PS w_MPB w_TSS_unfloc w_TSS_floc w_PIP_unfloc w_PIP_floc w_RDC w_RDN w_RDP w_LDP w_LDB nxsbs sedexch sinking rate of PL sinking rate of PS sinking rate of MPB sinking rate of TSS_unfloc sinking rate of TSS_floc sinking rate of PIP_unfloc sinking rate of PIP_floc sinking rate of RDC sinking rate of RDN sinking rate of RDP sinking rate of LDP sinking rate of LDB normalised excess bottom stress sediment exchange rate 0.5 m d-1 0.0 m d-1 5.0 m d-1 0.1 m d-1 5.0 m d-1 0.1 m d-1 5.0 m d-1 2.0 m d-1 2.0 m d-1 2.0 m d-1 3.0 m d-1 10.0 m d-1 0.005 10-9 m s-1

Technical Description of the CSIRO SERM Ecological Model

Constants The model uses a number of constants. Most of them relate to stoichiometry ratios (the Redfield ratio for planktonic plants photons:C:N:P:O2 = 1060:106:16:1:138; and the Atkinson ratio for benthic plants photons:C:N:P:O2 = 5500:550:30:1:716), although a few are physical constants. Constant BOLT PLANCK R g AV DNO3_25_0 DNH4_25_0 DPO4_25_0 red_A_I red_A_C red_A_N red_A_P red_A_O atk_A_I atk_A_C atk_A_N atk_A_P atk_A_O MW_Carb MW_Nitr MW_Phos MW_Oxyg red_W_C red_W_N red_W_P red_W_O atk_W_C atk_W_N atk_W_P atk_W_O C_O_W Description Boltzmann's constant Planck's constant Universal gas constant gravitational acceleration Avogadro's number Mol. diffusivity of nitrate in S = 0, T = 25 C Mol. diffusivity of ammonia in S = 0, T = 25 C Mol. diffusivity of phosphate in S = 0, T = 25 C Redfield molecular ratio photons:P ratio Redfield molecular ratio C:P ratio Redfield molecular ratio N:P ratio Redfield molecular ratio P:P ratio Redfield molecular ratio O2:P ratio Atkinson molecular ratio photons:P ratio Atkinson molecular ratio C:P ratio Atkinson molecular ratio N:P ratio Atkinson molecular ratio P:P ratio Atkinson molecular ratio O2:P ratio Molecular weight of carbon Molecular weight of nitrogen Molecular weight of phosphorus Molecular weight of oxygen Redfield weight ratio C:N Redfield weight ratio N:N Redfield weight ratio P:N Redfield weight ratio O2:N Atkinson weight ratio C:N Atkinson weight ratio N:N Atkinson weight ratio P:N Atkinson weight ratio O2:N Redfield /Atkinson weight ratio O2:C Value and units 1.38066 x 10-23 J K-1 6.62608 x 10-34 J s-1 8.31451 J K-1 mol-1 9.81 m s-2 6.02214 x 1023 mol-1 19.0 x 10-10 m2 s-1 19.8 x 10-10 m2 s-1 7.34 x 10-10 m2 s-1 1060.0 mol(quanta) mol(P)-1 106.0 mol(C) mol(P)-1 16.0 mol(N) mol(P)-1 1.0 mol(P) mol(P)-1 138.0 mol(O2) mol(P)-1 5500.0 mol(quanta) mol(P)-1 550.0 mol(C) mol(P)-1 30.0 mol(N) mol(P)-1 1.0 mol(P) mol(P)-1 716.0 mol(O2) mol(P)-1 12.01 g C mol-1 14.01 g N mol-1 30.97 g P mol-1 32.00 g O2 mol-1 5.68 g C g N-1 1.00 g N g N-1 0.138 g P g N-1 19.7 g O2 g N-1 15.7 g C g N-1 1.0 g N g N-1 0.074 g P g N-1 54.5 g O2 g N-1 3.469 g O2 g C-1

Technical Description of the CSIRO SERM Ecological Model

Ecological Processes The following list of processes are those appearing in the model equations. It should be noted that some processes are a made up of a series of other process (i.e. PLgrowth is made up of PLuptakeNH, PLuptakeNO and PLuptakeDIP). Such 'sub' processes are written in italics below the processes of which they are a component. Process PLgrowth PLuptakeDIC PLuptakeNH PLuptakeNO PLuptakeDIP PLreleaseO2 PLmortality Description the assimilation of dissolved inorganic compounds into large phytoplankton biomass uptake rate of carbon by large phytoplankton uptake rate of ammonia by large phytoplankton uptake rate of nitrate by large phytoplankton uptake rate of phosphorus by large phytoplankton release of O2 due to large phytoplankton growth Units mg N m-3 s-1 mg C m-3 s-1 mg N m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg O m-3 s-1

the loss rate of large phytoplankton to labile detritus at mg N m-3 s-1 the Redfield ratio (LDP) the assimilation of dissolved inorganic compounds into small phytoplankton biomass uptake rate of carbon by small phytoplankton uptake rate of ammonia by small phytoplankton uptake rate of nitrate by small phytoplankton uptake rate of phosphorus by small phytoplankton release of O2 due to small phytoplankton growth mg N m-3 s-1 mg C m-3 s-1 mg N m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg O m-3 s-1

PSgrowth PSuptakeDIC PSuptakeNH PSuptakeNO PSuptakeDIP PSreleaseO2 PSmortality

the loss rate of small phytoplankton to labile detritus mg N m-3 s-1 at the Redfield ratio (LDP) the assimilation of dissolved inorganic compounds into microphytobenthos biomass uptake rate of carbon by microphytobenthos uptake rate of ammonia by microphytobenthos uptake rate of nitrate by microphytobenthos uptake rate of phosphorus by microphytobenthos release of O2 due to microphytobenthos growth mg N m-3 s-1 mg C m-3 s-1 mg N m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg O m-3 s-1

MPBgrowth MPBuptakeDIC MPBuptakeNH MPBuptakeNO MPBuptakeDIP MPBreleaseO2 MPBmortality

the loss rate of small phytoplankton to labile detritus mg N m-3 s-1 at the Redfield ratio (LDP) assimilation of zooplankton small phytoplankton by small mg N m-3 s-1

ZSgrowth

Technical Description of the CSIRO SERM Ecological Model

ZSgrazePS ZSsloppyNH ZSsloppyDIP ZSsloppyDIC ZSsloppyO2

small phytoplankton loss due to grazing by small zooplankton ammonia release by small zooplankton due to inefficient (or sloppy) grazing phosphorus release by small zooplankton due to inefficient (or sloppy) grazing carbon release by small zooplankton due to inefficient (or sloppy) grazing oxygen consumption by bacteria during reminerialisation of small zooplankton sloppy grazing assimilation of large phytoplankton and microphytobenthos by large zooplankton large phytoplankton loss due to grazing by large zooplankton microphytobenthos loss due to grazing by large zooplankton ammonia release by large zooplankton due to inefficient (or sloppy) grazing phosphorus release by large zooplankton due to inefficient (or sloppy) grazing carbon release by large zooplankton due to inefficient (or sloppy) grazing oxygen consumption by bacteria during reminerialisation of large zooplankton sloppy grazing loss rate of large zooplankton production of labile detritus at the Redfield ratio (LDP) due to large zooplankton mortality release of ammonia through large zooplankton mortality release of phosphorus through large zooplankton mortality release of carbon through large zooplankton mortality oxygen consumption by bacteria during the remineralisation of large zooplankton loss rate of small zooplankton production of labile detritus at the Redfield ratio (LDP) due to small zooplankton mortality release of ammonia through small zooplankton mortality oxygen release by the bacterial during reminerialisation of small zooplankton sloppy grazing oxygen consumption by bacteria during the remineralisation of small zooplankton
9

mg N m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg C m-3 s-1 mg O m-3 s-1

ZLgrowth ZLgrazePL ZLgrazeMB ZLsloppyNH ZLsloppyDIP ZLsloppyDIC ZLsloppyO2

mg N m-3 s-1 mg N m-3 s-1 mg N m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg C m-3 s-1 mg O m-3 s-1 mg N m-3 s-1 mg N m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg C m-3 s-1 mg O m-3 s-1

ZLmortality ZLmortLDP ZLmortNH ZLmortDIP ZLmortDIC ZLmortO2

ZSmortality ZSmortLDP ZSmortNH ZSsloppyO2 ZSmortO2

mg N m-3 s-1 mg N m-3 s-1 mg N m-3 s-1 mg O m-3 s-1 mg O m-3 s-1

Technical Description of the CSIRO SERM Ecological Model

LDPbreak LDPtoNH LDPtoDIC LDPtoDIP LDPtoDOC LDPtoDON LDPtoDOP LDPtoRDC LDPtoRDN LDPtoRDP LDPconO2

the breakdown of labile detritus at the Redfield ratio remineralisation of labile detritus at the Redfield ratio to dissolved ammonia remineralisation of labile detritus at the Redfield ratio to dissolved inorganic carbon remineralisation of labile detritus at the Redfield ratio to dissolved inorganic phosphorus dissolution of labile detritus at the Redfield ratio into dissolved organic carbon dissolution of labile detritus at the Redfield ratio into dissolved organic nitrogen dissolution of labile detritus at the Redfield ratio into dissolved organic phosphorus production of refractory detrital carbon from the breakdown of labile detritus at the Redfield ratio production of refractory detrital nitrogen from the breakdown of labile detritus at the Redfield ratio production of refractory detrital phosphorus from the breakdown of labile detritus at the Redfield ratio consumption of O2 by bacterial remineralisation of LDP the breakdown of labile detritus at the Atkinson ratio remineralisation of labile detritus at the Atkinson ratio to dissolved ammonia remineralisation of labile detritus at the Atkinson ratio to dissolved inorganic carbon remineralisation of labile detritus at the Atkinson ratio to dissolved inorganic phosphorus dissolution of labile detritus at the Atkinson ratio into dissolved organic carbon dissolution of labile detritus at the Atkinson ratio into dissolved organic nitrogen dissolution of labile detritus at the Atkinson ratio into dissolved organic phosphorus production of refractory detrital carbon from the breakdown of labile detritus at the Atkinson ratio production of refractory detrital nitrogen from the breakdown of labile detritus at the Atkinson ratio production of refractory detrital phosphorus from the breakdown of labile detritus at the Atkinson ratio consumption of O2 by bacterial remineralisation of LDB

mg N m-3 s-1 mg C m-3 s-1 mg P m-3 s-1 mg C m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg C m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg O m-3 s-1 mg N m-3 s-1 mg N m-3 s-1 mg C m-3 s-1 mg P m-3 s-1 mg C m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg C m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg O m-3 s-1

LDBbreak LDBtoNH LDBtoDIC LDBtoDIP LDBtoDOC LDBtoDON LDBtoDOP LDBtoRDC LDBtoRDN LDBtoRDP LDBconO2

10

Technical Description of the CSIRO SERM Ecological Model

RDbreak

breakdown of refractory detritus (not explicitly in the model equations) remineralisation of refractory detrital carbon into dissolved inorganic carbon remineralisation of refractory detrital nitrogen into dissolved ammonia remineralisation of refractory detrital phosphorus into dissolved inorganic phosphorus dissolution of refractory detrital carbon into dissolved inorganic carbon dissolution of refractory detrital nitrogen into dissolved inorganic carbon dissolution of refractory detrital phosphorus into dissolved inorganic carbon consumption of O2 by bacterial remineralisation of RDC breakdown of dissolved organics (not explicitly in the model equations) remineralisation of dissolved organic carbon into dissolved inorganic carbon remineralisation of dissolved organic nitrogen into dissolved ammonia remineralisation of dissolved organic phosphorus into dissolved inorganic phosphorus consumption of O2 by bacterial remineralisation of DOC mg C m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg O m-3 s-1 mg C m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg C m-3 s-1 mg N m-3 s-1 mg P m-3 s-1 mg O m-3 s-1

RDCtoDIC RDNtoNH RDPtoDIP RDCtoDOC RDNtoDON RDPtoDOP RDCconO2

DOMbreak

DOCtoDIC DONtoNH DOPtoDIP DOCconO2

flocculate

the flocculation of total suspended solids (TSS) from kg m-3 s-1 the unflocculated TSS (TSS_unfloc) to the flocculated TSS (TSS_floc) the desorption (or, adsorption if negative) of mg P m-3 s-1 phosphorus from the particulate form (PIP) to the dissolved form (DIP) the transfer of PIP from PIP_unfloc to PIP_floc as a mg P m-3 s-1 result of the flocculation of TSS the assimilation of dissolved inorganic nitrogen, mg N m-2 s-1 phosphorus and carbon into macroalgae biomass uptake rate of carbon by macroalgae biomass mg C m-2 s-1 uptake rate of ammonia by macroalgae biomass mg N m-2 s-1

desorbedP

P_flocculate

MAgrowth MAuptakeDIC MAuptakeNH

11

Technical Description of the CSIRO SERM Ecological Model

MAuptakeNO MAuptakeDIP MAreleaseO2 MAmortality

uptake rate of nitrate by macroalgae biomass uptake rate of phosphorus by macroalgae biomass release of O2 due to macroalgae growth

mg N m-2 s-1 mg P m-2 s-1 mg O m-2 s-1

the loss rate of macroalgae biomass to labile detritus at mg N m-2 s-1 the Atkinson ratio the assimilation of dissolved inorganic nitrogen, phosphorus and carbon into seagrass biomass uptake rate of carbon by seagrass biomass uptake rate of ammonia by seagrass biomass uptake rate of nitrate by seagrass biomass uptake rate of phosphorus by seagrass biomass release of O2 due to seagrass growth mg N m-2 s-1 mg C m-2 s-1 mg N m-2 s-1 mg N m-2 s-1 mg P m-2 s-1 mg O m-2 s-1

SGgrowth SGuptakeDIC SGuptakeNH SGuptakeNO SGuptakeDIP SGreleaseO2 SGmortality

the loss rate of seagrass biomass to labile detritus at mg N m-2 s-1 the Atkinson ratio the conversion by bacteria of ammonia to nitrite and mg N m-3 s-1 then nitrate the conversion by bacteria of nitrate to N2 gas, which mg N m-3 s-1 is lost to the atmosphere

nitrification

denitrification

In addition to the ecological processes, physical processes such as light absorption, particle sinking and fluid mixing are modelled. Light absorption is discussed below, while particle sinking and fluid mixing are part of the physical model and are described elsewhere (Murray and Parslow, 1997).

12

Technical Description of the CSIRO SERM Ecological Model

Model Equations The model equations describe the change in the value of a state variable, as a sum of the effect of independent processes. The model equations have been group in water column, epibenthic and sediment categories. Epibenthic refers to processes occurring on the interface of the sediment and water column, and depending on both the state of the sediment and the water column. The equations have been written first in terms of the additional and subtraction of ecological processes. The quantification of these processes is given in the mathematical descriptions that follow in the next section. To apply these equations, a few points to watch for:

Epibenthic (EPI) processes have units per m2. To convert to a change in a water column state variable (which are specified per m3 of water) divide by water column layer thickness. To convert to a change in sediment state variable (which are specified per unit porewater volume) divide by sediment layer thickness and porosity. Epibenthic processes have the ability to impact on either water column (WC) or sediment (SED) state variables.

Water column equations: (dPL/dt)WC (dPS/dt)WC (dMB/dt)WC (dZL/dt)WC (dZS/dt)WC (dNH/dt)WC = PLgrowth - ZLgrazePL = PSgrowth - ZSgrazePS = MBgrowth - ZLgrazeMB = ZLgrowth - ZLmortality = ZSgrowth - ZSmortality = - PLuptakeNH - PSuptakeNH - MBuptakeNH + ZLsloppyNH + ZLmortNH + ZSsloppyNH + ZSmortNH + DONtoNH + LDPtoNH + LDBtoNH + RDNtoNH = - PLuptakeNO - PSuptakeNO - MBuptakeNO

(dNO/dt)WC

(dLDP/dt)WC = ZLmortLDP + ZSmortLDP - LDPbreak (dLDB/dt)WC = - LDBbreak (dRDC/dt)WC = LDPtoRDC + LDBtoRDC - RDCtoDIC - RDCtoDOC (dRDN/dt)WC = LDPtoRDN + LDBtoRDN - RDNtoNH - RDNtoDON (dRDP/dt)WC = LDPtoRDP + LDBtoRDP - RDPtoDIP - RDPtoDOP

13

Technical Description of the CSIRO SERM Ecological Model

(dDOC/dt)WC = LDPtoDOC + LDBtoDOC + RDCtoDOC - DOCtoDIC (dDON/dt)WC = LDPtoDON + LDBtoDON + RDNtoDON - DONtoNH (dDOP/dt)WC = LDPtoDOP + LDBtoDOP + RDPtoDOP - DOPtoDIP (dDO/dt)WC = + PLreleaseO2 + PSreleaseO2 + MBreleaseO2 - ZLmortO2 - ZLsloppyO2 - ZSmortO2 - ZSsloppyO2 - LDPconO2 - LDBconO2 - RDCconO2 + DOCconO2

(dDIC/dt)WC = -PLuptakeDIC - PSuptakeDIC - MBuptakeDIC + ZLmortDIC + ZLgrazeDIC + ZSmortDIC + ZSgrazeDIC + DOCtoDIC + LDPtoDIC + LDBtoDIC +RDCtoDIC (dTSS_unfloc/dt)WC = - flocculate (dTSS_floc/dt)WC (dDIP/dt)WC = + flocculate

= -PLuptakeDIP - PSuptakeDIP - MBuptakeDIP + ZLsloppyDIP + ZLmortDIP + ZSsloppyDIP + ZSmortDIP + DOPtoDIP + LDPtoDIP + LDBtoDIP + RDPtoDIP + desorbedP = -desorbedP_floc + P_flocculate = -desorbedP_unfloc - P_flocculate

(dPIP_floc/dt)WC (dPIP_unfloc/dt)WC In the epibenthos: (dMA/dt)EPI (dSG/dt)EPI (dNH/dt)WC (dNH/dt)SED (dNO/dt)WC (dNO/dt)SED

= MAgrowth - MAmortality = SGgrowth - SGmortality = - MAuptakeNH = - SGuptakeNH = - MAuptakeNO = - SGuptakeNO

(dLDB/dt)WC = + MAmortality (dLDB/dt)SED = + SGmortality (dO/dt)WC = + SGreleaseO2 + MAreleaseO2

14

Technical Description of the CSIRO SERM Ecological Model

(dDIP/dt)WC

= - MAuptakeDIP

(dDIP/dt)SED = - SGuptakeDIP (dDIC/dt)WC = - MAuptakeDIC (dDIC/dt)SED = - SGuptakeDIC In the sediment: (dPL/dt)SED (dPS/dt)SED = - PLmortality = - PSmortality

(dMB/dt)SED = MBgrowth - MBmortality (dNH/dt)SED = + LDPtoNH + LDBtoNH + DONtoNH + RDNtoNH - nitrification - MBuptakeNH = nitrification - denitrification - MBuptakeNO

(dNO/dt)SED

(dLDP/dt)SED = PLmortality + PSmortality + MBmortality - LDPbreak (dLDB/dt)SED = -LDBbreak (dRDC/dt)SED = LDPtoRDC + LDBtoRDC - RDCtoDIC - RDCtoDOC (dRDN/dt)SED = LDPtoRDN + LDBtoRDN - RDNtoNH - RDNtoDON (dRDP/dt)SED = LDPtoRDP + LDBtoRDP - RDPtoDIP - RDPtoDOP (dDOC/dt)SED = LDPtoDOC + LDBtoDOC + RDCtoDOC - DOCtoDIC (dDON/dt)SED = LDPtoDON + LDBtoDON + RDNtoDON - DONtoNH (dDOP/dt)SED = LDPtoDOP + LDBtoDOP + RDPtoDOP - DOPtoDIP (dDO/dt)SED = + MBreleaseO2 - LDPconsO2 - LDBconsO2 - RDCconsO2 - DOCconsO2

(dDIP/dt)SED = - MBuptakeDIP + DOPtoDIP + LDPtoDIP + LDBtoDIP+ RDPtoDIP + desorbedP (dPIP_floc/dt)SED = -desorbedP_floc

15

Technical Description of the CSIRO SERM Ecological Model

(dPIP_unfloc/dt)SED

= -desorbedP_unfloc

(dDIC/dt)SED = - MBuptakeDIC + DOCtoDIC + LDPtoDIC + LDBtoDIC + RDCtoDIC

16

Technical Description of the CSIRO SERM Ecological Model

Physical Processes Description The ecological code used by SERM contains the physical processes of light attenuation and the temperature dependence of ecological processes, which are discussed below. The physical code contains particulate sinking, which, as it directly impacts the ecological state variables, is also discussed below. The transport of ecological tracers by the physical code is dealt with in the description of the physical code. Light attenuation. In the model, light is attenuated through the water column, the benthic macroalgae, the seagrass, and the microphytobenthos sequentially. Photosynthetically available radiation (PAR) at the bottom of a layer of water, Ibot [mol photon m-2 s-1], is given by: Ibot = Itopexp(-Kd dz) where Itop is the PAR at the top of the layer [mol photon m-2 s-1], dz is the thickness of the layer [m], and Kd is the total attenuation coefficient of the water [m-1]. Kd is given by the sum of the each attenuating component in the water: Kd = kW + nPS aAPS + nPL aAPL + nMPB aAMPB + kDON DON + kDL(LD + RDN) + kTSS TSS + KCDOM (35 - S) / 35 where kw is the background attenuation coefficient of water [m-1], nPS, aAPS, nPL, aAPL and nMPB and aAMPB are the concentration [cell m-3] and absorption cross-section [m2 cell-1] of cells of small phytoplankton, large phytoplankton and microphytobenthos respectively, kDON and kDL are the nitrogen-specific attenuation coefficients of the dissolved organic matter and labile detritus [mg N-1 m2], kTSS is the mass specific attenuation coefficient of total suspended solids [kg-1 m2]. LD, RDN and DON are the concentrations of biomass of labile detritus, refractory detritus and dissolved organic matter [mg N m-3], and TSS is the concentration of total suspended solids [kg m-3]. KCDOM is the attenuation coefficient of the non-living constituents of fresh water flowing into the estuary [m-1] and S is salinity [PSU]. The average irradiance in the layer, Iav [mol photon m-2 s-1], is given by: Iav = (Itop - Ibot) / (Kd dz) In the model, light reaching the benthos is first attenuated by macroalgae. The light below the macroalgae is given by: I below MA = Ibot exp(-MA aAMA) where MA is the biomass of macroalgae [mg N m-2] and aAMA is the biomass-specific absorption cross-section [mg N-1 m2]. The light below the seagrass is given by: Ibelow SG = Ibelow MA exp(-SG aASG)

17

Technical Description of the CSIRO SERM Ecological Model

where SG is the biomass of seagrass [mg N m-2] and aASG is the biomass-specific absorption cross-section [mg N-1 m2]. The PPBES ecological model considered the reduced growth rate of seagrass due to epiphytes. In this paper, we have considered epiphytes to be part of the benthic macroalgae. Epiphytes and macroalgae both receive light before seagrass and, sharing the same surface, have a similar effective benthic boundary layer thickness. As epiphytes and macroalgae have similar maximum supply rates of nutrient and light, and shade the seagrass in a similar manner, the assumption they make up one class of autotroph seemed reasonable3. Finally, the remaining light passes through the microphytobenthos at the surface of the sediment: Ibelow MPB = Ibelow SG exp(-nMPB aAMPB dz) where nMPB is the concentration of microphytobenthos cells [cell m-3] with an absorption crosssection of aAMPB [m2 cell-1] in a sediment layer dz thick [m]. By including only attenuation due to microphytobenthos we are assuming that the microphytobenthos lie in the surface layer of the sediment. The average light flux available to the microphytobenthos cells is given by: Iav = (Ibelow SG - Ibelow:MPB) / (nMPB aAMPB dz) The above description of the light field in an estuarine environment is significantly different to that employed in the PPBES ecological model, and other aquatic ecological models (Fasham, 1990, Madden et. al., 1996). Autotroph absorption cross-sections have been used to parameterise both the dependence of autotroph growth rate on light availability and the attenuation of light as it passes through the water column and benthos. In all, ten parameters with well defined physical interpretations were used to model the effect of light on ecological processes. The PPBES ecological model contained the physically well-defined kw, kDON, kDL (and in principle kTSS and kCDOM), but also contained five empirically-determined half-saturation constants relating growth rate to incident light, and two parameters to describe the relationship between light attenuation and chlorophyll concentration (twelve parameters in all). The additional two parameters were required by the PPBES model because phytoplankton growth and the attenuation rate of light were considered independent processes. In contrast, the above presented mechanistic model takes advantage of the common physical process of light capture by cells that determines both light attenuation through the water column and phytoplankton photosynthesis. Sinking Phytoplankton cells (large and small), microphytobenthos, and detritus (refractory and labile) often have a different density than water. Their resultant change in concentration of particle X, [X], from a water column layer as a result of sinking is given by: d [X] /dt = - w_X [X] / layer thickness where w_X is the sinking rate [m d-1].
3

A problem with this assumption is that epiphytes have a Redfield C:N:P ratio, and probably a maximum growth rate closer to suspended algae than macroalgae. Nonetheless, for the benefit of simplicity, the assumption that epiphytes and macroalgae are modelled as one autotrophic class was used.

18

Technical Description of the CSIRO SERM Ecological Model

Temperature dependence A number of ecological processes are temperature dependent. To capture this, the relevant biological parameters have been multiplied by: Tcorr = Q10(T-Tref)/10 where Q10 is the temperature coefficient for rate parameters [dimensionless], and Tref is the reference temperature [C]. Biological parameters with a temperature dependence are: mL_PL_T15, mL_MA_T15, mL_SG_T15, mQ_MB_T15, mQ_ZL_T15, mQ_ZS_T15, mS_MA_T15, mS_SG_T15, r_LDP_T15, r_LDB_T15, r_RD_T15, r_DOM_T15, PLumax, PSumax, MBumax, ZSumax, ZLumax, ZSswim, Zlswim. Temperature, T, (and salinity, S) have a secondary effect on some ecological processes (particularly grazing) by changing the viscosity of water. The changing the viscosity of water and the molecular diffusivity of chemical species with temperature and salinity is modelled following Wolf-Gladrow et al. (1997): viscosity (T, S = 0) = 0.00123161 - 1.228e-5 T + 4.009e-8 T2 viscosity (T, S = 35) = viscosity (T, S = 0) /(0.9508-0.0007379*T) viscosity(T, S) = viscosity (T, S = 35) *(S/35) + viscosity (T, S = 0) *(35-S)/35 For molecular diffusivity: D (T, S) = D25S0 * viscosity(T, S) * (T+273.15)/(viscosity (T, S = 0) *298.15)

19

Technical Description of the CSIRO SERM Ecological Model

Ecological Process Description Primary Production (plant growth) Primary productivity (the organic synthesis from inorganic substances) is determined for each autotroph (large and small phytoplankton, microphytobenthos, macroalgae and seagrass) from a functional form specifying the interaction of the maximum supply rates of nutrients and light, and the maximum growth rate. Note that this is fundamentally different to the typical Monod or Michaelis-Menton type growth functions, which use half-saturation constants. This scheme (called the CR model in Baird and Emsley, 1999) is preferred because maximum rates can be mechanistically determined. The CR model requires the determination of the maximum growth rate, m [s-1], and the maximum supply rates of nutrients, kN [mol N s-1], and light, kI [mol (quanta) m-2 s-1], and the stoichiometry of each autotroph [units vary - see below]. The determination of maximum supply rates depends on morphology and location of the autotroph. The following considers phytoplankton (suspended algal cells), benthic microalgae (algal cells in the surface sediment layer), macroalgae attached to the bottom, and seagrass. Maximum uptake rates of suspended algal cells. Suspended algal cells (phytoplankton) obtain nutrients from the water column, and light as a function of the average light in the layer in which they are suspended. Biomass is quantified per m3. The maximum supply rates of nutrients (limited by molecular diffusion to a perfectly absorbing cell) is given by: k N = DN where is the diffusion shape factor [m]; D is the molecular diffusivity of the nutrient [m2 s-1], and N is the concentration of the nutrient in the water column [mol m-3]. for a sphere is 4r, where r is the radius [m]. The diffusion shape factor for more complicated shapes (spheroids, cubes etc.) can be found in Baird and Emsley (1999) or Baird (1999). The maximum supply rate of light is given by: k I = I av aA cell where aAcell is the absorption cross section of a cell [m2]; and Iav is the average PAR radiation [mol (photons) m-2 s-1] in the layer. The absorption cross section of a sphere is given by: 2 1 1 + 2Cr e 2Cr 2 aAcell = r 1 2 2Cr

( (

where C [m-1] is the absorption coefficient. For a black cell, 2Cr approaches , and aAcell approaches r2 and for a non-absorbing cell, 2Cr approaches 0, and aAcell approaches 0. The absorption cross section more complicated shapes can be found in Baird and Emsley (1999) and Baird (1999). Maximum uptake rates of benthic macroalgae. Benthic macroalgae reside on the top of the sediment, take nutrients out of the water column, and are exposed to light that reaches the bottom of the water column. Biomass is quantified per m2.

20

Technical Description of the CSIRO SERM Ecological Model

The maximum supply rate of nutrients is strongly influenced by shear stress at the water/sediment interface. The effect of this on nutrient uptake has been quantified by analogy with an experimental investigation of heat transfer, surface roughness and pressure drop in pipes. The application of this analogy to heat transfer to benthic communities has been demonstrated (Baird and Atkinson, 1997, Thomas et al., 2000). First a friction velocity, U* [m s-1] (a characteristic velocity used to represent the velocity within the momentum boundary layer) is determined: U * = Ub c f / 2 where Ub is the velocity of the fluid away from the boundary [m s-1], cf is the friction coefficient, a dimensionless measure of the shear stress exerted on a fluid by the surface, arising from both skin friction and form drag. Then a Reynold's roughness number, Rek, a non-dimensional measure of the surface roughness is determined: U *k s Re k = where ks is an equivalent (in creating friction) sand-grain roughness dimension [m], and is the kinematic viscosity [m2 s-1]. The Stanton number, Stm [-], the ratio of the uptake of a substance to the advection of that substance past the surface, can be calculated from: Stm = (c f 2) 0.9 + c f 2 Stk1

where Stk = 5.19 Re0.2 Sc 0.44 8.48 k and Sc is the Schmidt number, (Sc = /D), the ratio of molecular diffusivity of momentum to the molecular diffusivity of mass. From the Stanton number, the maximum uptake rate, kN, [mol N m-2] is given by: k N = St mU b N where N is the nutrient concentration in the water column [mol N m-3]. As discussed later, StmUb = D / , where is the benthic diffusive boundary layer thickness, and is a function of , D, temperature and ks, and is of the order of 0.1 mm for a typical benthic macrophyte. The maximum light capture by the macroalgae is given by: k I = I 1 exp( MAaA MA where I is the incident radiation at the top of the macroalgae [mol(photons) m-2 s-1]; MA is the biomass of macroalgae [mg N m-2]; and aA MA is the nitrogen-specific absorption cross-section of macroalgae [m2 mg N-1].

Maximum uptake rates of seagrass. Seagrass obtain their nutrients from the sediment, and their light after it has passed through the water column and benthic macroalgae. For seagrass, a maximum supply rate of nutrients through the root system is difficult to quantify. Instead, a maximum supply rate can be "back-calculated" from a half-saturation and a maximum growth rate, using

21

Technical Description of the CSIRO SERM Ecological Model

kN = SG m N /Km where SG is the biomass of seagrass [mol N m-2]; m is the maximum growth rate [s-1], N is the sediment pore-water nutrient concentration [mol N m-3], and Km [mol N m-3] is an experimentally determined half-saturation constant of growth when fitted to the Monod growth equation. The implication of the last equation is that the rate of nutrient uptake increases with biomass (like phytoplankton cells), but unlike macroalgae (which has a constant uptake per m2). The increase in uptake with biomass is probably realistic, as a higher biomass of seagrass would have a larger root system (in the same way that a higher biomass of phytoplankton cells would have a larger surface area). Maximum light uptake by seagrass is given by: k I = I 1 exp( SG * aA SG where I is the light that has passed through the macroalgae; SG is the biomass of seagrass [mol N m-2]; and aA SG is the nitrogen-specific absorption cross-section of seagrass [m2 mg N-1].

Maximum uptake rates of benthic microalgae. Benthic microalgae exist in the top layer of the sediment. The maximum nutrient uptake (from the sediment porewater) is the same as for suspended algal cells (although the diffusion coefficient may be reduced). The maximum light capture is given by: k I = I 1 exp(nMB aA MB / nMB where I is the light that has passed through the water column, macroalgae and seagrass [mol (photons) m-2 s-1], nMB is the concentration of cells [cell m-3]; aA MB is the absorption crosssection of the benthic microalgae [m2].

Growth rates of autotrophs We now have the maximum uptake rates of light and nutrients for a number of autotrophs classes. How do we combine these rates and the maximum growth rate to obtain a prediction of the growth rate at a particular nutrient concentration and irradiance? We are not able to continue our strict mechanistic derivations to intracellular processes, which are both less well understood and far more complex than the extracellular processes described above. To combine the rates we will make two assumptions (which are more fully explored in Baird et. al., 2001) to guide our final form. In essence, though, we are now pursuing an empirical approach. The uptake rate of nutrients and light (or at least fixed carbon, which is the storage medium for energy) is a function of the resource already stored. For a light and nitrate limited system: Uptake of N = kN(1-RN) Uptake of I = kI(1-RI) where R is a measure, between 0 and 1, or the reserves of a nutrient available for growth. Now assume that growth rate [s-1] is determined by a linear product of R for both light and nitrate:
= m . RN . RI

22

Technical Description of the CSIRO SERM Ecological Model

The balance between growth and uptake becomes: kI(1-RI) = m .RI .RI mI kN(1-RI) = m .RN . RI mN where mN and mI are stoichiometry coefficients, or the number of moles of N, I per unit of biomass (either cell or m-2). I will call this the benthic solution. Because of the convention of representing C:N:P ratio relative to P, it is easiest to define a mP = 9140*(4/3)r3/106 (Baird,1999), and use the Redfield ratio of 848:106:16:1 of photons:C:N:P to obtain mI, mC, and mN. For cells that divide (and therefore have to share luxury resource between cells), the solution becomes: kI(1-RI) = m .RN .RI (mI + RIRImax) kN(1-RN) = m .RN .RI (mN + RNRNmax) To obtain a growth rate (from = m . RN . RI ) requires knowledge of RN and RI. Ideally in an ecological model, RN and RI will be state variables, and their values are tracked through time (as in Baird and Emsley, 1999). However, with a number of algal and seagrass species, this may be too time consuming. Instead, if it is assumed that the autotroph has reached an equilibrium growth state, RN and RI can be determined at every time point if m, mI, mN, kN and kI are known. The equilibrium state is obtained by solving the two non-linear simultaneous equations by applying Newton's method for solving systems of non-linear equations, truncating the Taylor series approximation to one term, and using Gaussian elimination to solve the intermediate linear simultaneous equations, until successive approximations were within 10-9 (Baird and Emsley, 1999). Again, this is a time consuming procedure. However, the solution for a range of supply rates can be placed in a look-up table, which reduces the computation time. Autotrophic Respiration Respiration is the release of energy from organic compounds. More complex models of autotrophic growth include respiration explicitly. In the SERM model, respiration is accounted for in two ways. The photon requirement for growth relative to P, red_A_I or atk_A_I is elevated from the theoretical minimum of 8 to 10. Secondly, the first Presp or Bresp (which were set at 0.025 for SERM) fraction of light that is required for the maximum growth rate are not counted towards growth. The increase from 8 to 10 is to account for the respiration during growth, while Presp and Bresp ensure a growth is only greater than respiration at a non-zero light level. Secondary Production (animal growth) The only secondary production explicitly modelled is zooplankton growth. Zooplankton cells are only found in the water column and consume only phytoplankton. The model contains of two

23

Technical Description of the CSIRO SERM Ecological Model

classes of zooplankton, small (ZS) and large (ZL), which graze on the small and large (both MB and PL) phytoplankton respectively. Zooplankton growth is calculated from the encounter rates of phytoplankton and zooplankton cells (Jackson, 1996). If the encounter rate exceeds maximum growth rate of the zooplankton cell, the grazing rate is set to the encounter rate. To go through the calculation for small zooplankton. Encounter rates are based on number of cells per unit volume, so a conversion between biomass [mg N m-3] and cell numbers [cell m-3] is required for both predator (ZS) and prey species (PS): ZScells = ZS/(ZSm*16*14010) PScells = PS/(PSm*16*14010) The encounter rate between one predator and a population of prey cells [cell (PS) m-3] is given by: encounter_rate_ZS = PScells*ZSphi_PS [cell (PS) s-1] where ZSphi_PS [m3 s-1] is the encounter rate coefficient between species ZS and PS. The encounter rate coefficient is based on the encounter rate of spheres by summing three processes that bring predators and prey into contact: diffusion, relative motion (swimming and sinking) and fluid shear: phi = phidiff + phirelmotion + phishear The diffusive encounter rate is calculated using: phidiff = (2.0*BOLT*Twater/(3.0*density*dyn_viscosity))*(1/PSrad+1/ZSrad)*(PSrad+ZSrad) where PSrad is the radius of small phtyoplankton [m], ZSrad is the radius of small zooplankton [m], viscosity is the dynamic viscosity of the water [kg s-1 m-1], epsilon is the dissipation of turbulent kinetic energy [m2 s-3], density is the density of water [kg m-3] and BOLT is Boltzmann's constant [J K-1] and Twater is the temperature of the water [C]. The relative motion encounter rate is given by: phirelmotion = (PSrad+ZSrad)2*Ueff where Ueff is the effective relative motion between PS and ZS [m s-1]. Relative motion (other than resulting from diffusion and shear) consists of two primary sources: the relative sinking velocities of the two cells, and the swimming velocities of the two cells. In the NLWRA application, we have included only the swimming velocities, which are combined to an effective relative velocity by: U eff =
2 U slow + 3U 2 fast

3U fast where Uslow is the slower of PSswim and ZSswim, and Ufast is the faster of PSswim and ZSswim. PSswim [m s-1] has been determined from a general size-based relationship:

24

Technical Description of the CSIRO SERM Ecological Model

PSswim = 0.004 PSrad0.26 while ZSswim is specified as a parameter. The encounter rate coefficient due fluid shear is given by: phishear = 1.3*(epsilon/kin_viscosity)0.5*(PSrad+ZSrad)3 where PSswim is the swimming velocity of the prey [m s-1], ZSswim is the swimming velocity of the predator [m s-1] and kin_viscosity is the kinematic viscosity [m2 s-1]. The maximum ingestion rate, however, is limited by how fast the predator can grow. The maximum growth rate is in units of s-1, and must be converted to cells of prey ingested to maintain the maximum growth rate [cell (PS) cell (ZS)-1 s-1] max_ingestion_rate_ZS = ZSumax*(ZSm/PSm)/E_ZS where E_ZS is the efficiency of small zooplankton growth. The clearance rate of PS cells by ZS cells becomes the minimum of the encounter rate and the maximum ingestion rate per ZS cells multiplied by the number of ZS cells: ZS_CLEAR = min(max_ingestion_rate_ZS, encounter_rate_ZS)*ZScells We know need to convert back to units of biomass for grazing loss of PS, ZSgrazePS, and the subsequent gain in biomass of ZS (account for an efficiency loss) ZSgrazePS = ZS_CLEAR / (PSm*16*14010)) ZSgrowth = E_ZS * ZSgrazePS The same procedure is undertaken for large zooplankton, ZL, except that large zooplankton feed on two types of algal cells, large phytoplankton and suspended microphytobenthos. In this case, it was chosen to have no feeding preference between the two prey types, so that their loss rates was always in proportion to their encounter rates (even if the large zooplankton were ingestion rate limited). Mortality (unmodelled trophic level effects) A mortality rate is specified for a number of the plants and animals to represent loss processes which are not already considered. Given that we consider physical loss terms (sinking and advection) in the physical model, and grazing loss on phytoplankton by zooplankton, mortality essentially becomes the effect of unparameterised trophic levels (such as viruses, fish larvae, whales etc.) and trophic interactions (bacterial decay). The mortality terms in the model are: In the water column: The only mortality terms in the water column are loss rates of zooplankton:

25

Technical Description of the CSIRO SERM Ecological Model

ZLmortality = mQ_ZL * ZL * ZL ZSmortality = mQ_ZS * ZS * ZS A quadratic form was chosen to represent the effect of higher trophic levels. As zooplankton increase in numbers, their predators are expected to increase as well, and the combined effect may be best captured using a quadratic form. In the epibenthos Both seagrass and macroalgae are given mortality rates, primarily to capture the effect of grazing. MAmortality = mL_MA * MA SGmortality = mL_SG * SG In both cases, a linear loss rate was assumed. In the sediment Loss rates are given for phytoplankton cells (linear to represent bacterial decay of unviable cells), and microphytobenthos (quadratic to represent a higher trophic level): MAmortality = mL_MA * MA SGmortality = mL_SG * SG Total Suspended Solids flocculation We consider two classes of suspended solids: flocculated (TSS_floc) [kg m-3] and unflocculated (TSS_unfloc) [kg m-3], with sinking rates of w_floc [m s-1] and w_unfloc [m s-1] respectively. The rate at which TSS flocculates [kg m-3 d-1] (i.e. moves from TSS_unfloc to TSS_floc) is a discontinuous function of the salinity, S, and a parameter giving the maximum flocculation rate, r_floc [s-1]: S > 10.0 6.0 < S < 10.0 S < 6.0 flocculate = TSS_unfloc*r_floc flocculate = TSS_unfloc*r_floc * (S-6.0)/4.0 flocculate = 0

Additional factors influencing flocculation rate, but in the ecological model, include a dependence of TSS concentration, particle size distribution and shear rates. Phosphate absorption / desorption reactions In the water column, the reversible absorption / desorption reactions for PIP_floc and PIP_unfloc are described by:

26

Technical Description of the CSIRO SERM Ecological Model

-d(PIP_floc)/dt = d(DIP)/dt = [PIP_floc - P_abs_coef*TSS*DIP]*r_abs_P -d(PIP_unfloc)/dt = d(DIP)/dt = [PIP_unfloc - P_abs_coef*TSS*DIP]*r_abs_P P_abs_coef is the phosphate absorption coefficient [L g-1 or m3 kg-1], r_abs_P [s-1] is the rate at which absorption / desorption equilibrium is reach, and: TSS = TSS_floc + TSS_unfloc PIP = PIP_floc + PIP_unfloc In the water column, PIP moves from PIP_unfloc to PIP_floc at the same flocculation rate as TSS_unfloc moves to TSS_floc. In the sediment, we assume a constant quantity of TSS. Given the same P_abs_coef, a porosity of = 0.547 and a density of sediment of = 2650 kg m-3, a constant sediment buffering capacity, P_buff_sed, is given by: P_buff_sed = (1-) KD = (1-0.547)*2*2650 = 2,400 kg kg-1 The absorption / desorption reaction becomes independent of TSS, and is given by: -d(PIP_floc)/dt = d(DIP)/dt = [PIP_floc/P_buff_sed- DIP]*r_abs_P -d(PIP_unfloc)/dt = d(DIP)/dt = [PIP_unfloc/P_buff_sed- DIP]*r_abs_P Detritus (organic suspended particles) and dissolved organic matter processes Detrital and organic matter processes are represented the same in both the water column and sediment, although nitrification / denitrification occurs only in the sediment (see below). Nonliving organic matter is represented in three major components: labile detritus (fast breaking down organic particles), refractory detritus (slow breaking down organic particles) and dissolved organic matter (dissolved organic molecules). Labile detritus is composed of two classes based on the stoichiometry of the elements making up the labile detritus: one at the Redfield ratio (C:N:P 106:16:1) (LDP), and one at the Atkinson ratio (C:N:P 550:30:1) (LDB). LDP is typical of aquatically-derived plant detritus, while LDB is typical of terrestrial plants, and seagrasses. Refractory detritus and dissolved organic matter are composed of three pools each: a carbon (RDC and DOC), a nitrogen (RDN and DON) and phosphorus (RDP and DOP) [see figure]. Rates of decay The rates of detrital and dissolved organic matter decay are specified by breakdown rates of r_LDP, r_LDB, r_RD and r_DOM. For refractory detritus and dissolved organic matter, which are sub-divided into carbon, nitrogen and phosphorus pools, the breakdown rates of all pools are equal. Destination of decay material

27

Technical Description of the CSIRO SERM Ecological Model

As well as quantifying the rates of decay, we need to specify the destination of the decayed material. To do this, we represent the fraction of the decaying matter reaching equal breakdown pool. The fractions for labile detritus are specified as the fraction of labile detritus (both LDP and LDB) breaking down to refractory detritus, F_LD_RD, and dissolved organic matter, F_LD_DOM. The remaining fraction, (1- F_LD_RD - F_LD_DOM), is broken down to dissolved inorganic pools (DIC, NH and DIP). For refractory detritus (carbon, phosphorus and nitrogen pools), the fraction breaking down to dissolved organic matter is given by F_RD_DOM, while the remaining is broken down to dissolved inorganic pools (DIC, NH and DIP). The above listed detrital and organic matter breakdown processes can be written follows: The breakdown of LDP is given by: LDPbreak = r_LDP * LDP where the component going to each dissolved organic matter pool is given by: LDPtoDOC = r_LDP * LDP * F_LD_DOM * red_W_C LDPtoDON = r_LDP * LDP * F_LD_DOM LDPtoDOP = r_LDP * LDP * F_LD_DOM * red_W_P and the component going to each refractory detritus pool by: LDPtoRDC = r_LDP * LDP * F_LD_RD * red_W_C LDPtoRDN = r_LDP * LDP * F_LD_RD LDPtoRDP = r_LDP * LDP * F_LD_RD * red_W_P leaving the remainder to go into dissolved inorganic nutrients pools: LDPtoNH = r_LDP * LDP * (1.0 - F_LD_RD - F_LD_DOM) LDPtoDIC = r_LDP * LDP * (1.0 - F_LD_RD - F_LD_DOM) * red_W_C LDPtoDIP = r_LDP * LDP * (1.0 - F_LD_RD - F_LD_DOM) * red_W_P with the fraction going into dissolved inorganic nutrients consumes oxygen at a rate: LDPconO2 = red_W_O * r_LDP * LDP * (1.0 - F_LD_RD - F_LD_DOM) A similar set of terms for the breakdown of LDB are given by: LDBbreak = r_LDB * LDB where the component going to each dissolved organic matter pool is given by:

28

Technical Description of the CSIRO SERM Ecological Model

LDBtoDOC = r_LDB * LDB * F_LD_DOM * red_W_C LDBtoDON = r_LDB * LDB * F_LD_DOM LDBtoDOP = r_LDB * LDB * F_LD_DOM * red_W_P and the component going to each refractory detritus pool by: LDBtoRDC = r_LDB * LDB * F_LD_RD * red_W_C LDBtoRDN = r_LDB * LDB * F_LD_RD LDBtoRDP = r_LDB * LDB * F_LD_RD * red_W_P leaving the remainder to go into dissolved inorganic nutrients pools: LDBtoNH = r_LDB * LDB * (1.0 - F_LD_RD - F_LD_DOM) LDBtoDIC = r_LDB * LDB * (1.0 - F_LD_RD - F_LD_DOM) * red_W_C LDBtoDIP = r_LDB * LDB * (1.0 - F_LD_RD - F_LD_DOM) * red_W_P with the fraction going into dissolved inorganic nutrients consumes oxygen at a rate: LDBconO2 = red_W_O * r_LDB * LDB * (1.0 - F_LD_RD - F_LD_DOM) The breakdown of each pool of refractory detritus is given by: RDxbreak = RDx * r_RD where RDx represents each of the pools (carbon, nitrogen and phosphorus) of refractory detritus. The resulting breakdown of refractory to dissolved organic matter is given by: RDCtoDOC = RDC * r_RD * F_RD_DOM RDNtoDON = RDN * r_RD* F_RD_DOM RDPtoDOP = RDP * r_RD* F_RD_DOM with the remainder going to dissolved inorganic nutrient pools: RDCtoDIC = RDC * r_RD * (1 - F_RD_DOM) RDNtoNH = RDN * r_RD * (1 - F_RD_DOM) RDPtoDIP = RDP * r_RD * (1 - F_RD_DOM) with the consumption of O2 assumed to be a function of the carbon fraction of RD: RDCconO2 = RDC * r_RD * (1 - F_RD_DOM) * C_O_W The breakdown of each pool of dissolved organic matter is given by: DOxbreak = r_DOM * DOx

29

Technical Description of the CSIRO SERM Ecological Model

where DOx represents each of the pools (carbon, nitrogen and phosphorus) of dissolved organic matter. The resulting breakdown of dissolved organic matter to dissolved inorganic nutrients is given by: DOCtoDIC = DOC * r_DOM DONtoNH = DON * r_DOM DOPtoDIP = DOP * r_DOM with the consumption of O2 assumed to be a function of the carbon fraction of DOM: DOCconO2 = DOC * r_DOM * C_O_W Sediment Nitrification and Denitrification In the sediment, ammonia is subject to conversion to nitrite and then nitrate by bacteria (nitrification). A different group of bacteria are capable of converting nitrate to nitrite and then nitrogen gas (denitrification). These interacting processes have not been modelled individually, but instead their combined effect has been empirically approximated (see PPBES for more details). The empirical approximation is based on evidence that increased respiration decreases the loss of nitrogen to N2 gas. The loss of efficiency is thought to be primarily in the nitrification step (rather than the final denitrification step). The net respiration rate is given by: respiration = LDPtoNH + LDBtoNH + DONtoNH + RDNtoNH - MBgrowth where respiration is an areal flux (so must be corrected for sediment thickness). The amount of nitrification is given by: nitrification = respiration * Nitrific_eff where a change in nitrification efficiency, Nitrific_eff, is assumed to account for the reduction in the overall conversion of NH to N2. The empirical approximation for is Nitrific_eff : Nitrific_eff = Nitrific_eff_max * emax(1.0 - respiration / R_0) Denitrification rate is then given by: Denitrification = Nitrification * Denitrific_eff where Dentrific_eff represents the decrease in denitrification rate at very low respiration rates (below R_D): Denitrific_eff = emin(respiration / R_D, 1.0)

30

Technical Description of the CSIRO SERM Ecological Model

The percent of the ammonia that is respired which is released to the atmosphere as N2 gas is therefore given by: PercentDenitri = 100 * Denitrific_eff * Nitrific_eff

31

Technical Description of the CSIRO SERM Ecological Model

References Baird, M. E. (1999) Towards a verified mechanistic model of plankton population dynamics. Ph.D. thesis, University of Warwick. Download: http://www.oikos.warwick.ac.uk/ecosystems/ThesisArchive/baird thesis.html. Baird, M. E. and Atkinson, M. J. (1997) Measurement and prediction of mass transfer to experimental coral reef communities. Limnol. Oceanogr., 42, 1685-1693. Baird, M. E. and Emsley, S. M. (1999) Towards a mechanistic model of plankton population dynamics. J. Plankton Res., 21, 85-126. Baird, M. E., Emsley, S. M. and McGlade, J. M. (2001) Modelling the interacting effects of nutrient uptake, light capture and temperature on phytoplankton growth. J. Plankton Res. 23: 829-840. Bilger, R. W. and Atkinson, M. J. (1992) Anomalous mass transfer of phosphate on coral reef flats. Limnol. Oceanogr., 37, 261-272. Boudreau, B. P. (1996) The diffusive tortuosity of fine grained unlithified sediments. Geochim. Cosmochim. Acta, 60, 3139-3142. Bricker, S. B., Clement, C. G., Pirhalla, D. E., Orlando, S. P. and Farrow, D. R. G. (1999) National estuarine eutrophication assessment: Effects of nutrient enrichment in the nation's estuaries. Tech. rep., NOAA, National Ocean Service, Special Projects Office for Coastal Ocean Science, Silver Spring, MD. Chorus, I. and Bartram, J. (editors) (1999) Toxic cyanobacteria in water . E & FN Spon. Dennison, W. C. and Abal, E. G. (1996) Moreton Bay Study: a scientific basis for the healthy waterways campaign. South East Queensland Regional Water Quality Management Strategy, Brisbane, Australia. Dipprey, D. F. and Sabersky, D. H. (1963) Heat and momentum transfer in smooth and rough tubes at various Prandtl numbers. Int. J. Heat Mass Transfer , 6, 329-353. Enriquez, S., Agusti, S. and Duarte, C. M. (1994) Light absorption by marine macrophytes. Oecologia, 98, 121-129. Falkowski, P. G. (2000) Rationalizing elemental ratios in unicellular algae. J. Phycol , 36, 3-6. Fasham, M. J. R., Ducklow, H. W. and McKelvie, S. M. (1990) A nitrogen-based model of plankton dynamics in the oceanic mixed layer. J. Mar. Res., 48, 591-639.

32

Technical Description of the CSIRO SERM Ecological Model

Harris, G. (1999) Comparison of the biogeochemistry of lakes and estuaries: ecosystem processes, functional groups, hysteresis effects and interactions between macro- and microbiology. Mar. Fresh. Res., 50, 791-811. Harris, G., Batley, G., Fox, D., Hall, D., Jernakoff, P., Molloy, R., Murray, A., Newell, B., Parslow, J., Skyring, G. and Walker, S. (1996) Port Phillip Bay Environmental Study Final Report . CSIRO, Canberra, Australia. Healey, F. P. (1980) Slope of the Monod equation as an indicator of advantage in nutrient competition. Microb. Ecol., 5, 281-286. Hofmann, M., Wolf-Gladrow, D. A., Takahashi, T., Sutherland, S. C., Six, K. D. and MaierReimer, E. (2000) Stable carbon isotope distribution of particulate organic matter in the ocean: a model study. Mar. Chem., 72, 131-150. Hurd, C. L. (2000) Water motion, marine macroalgal physiology, and production. J. Phycol., 36, 453-472. Kirk, J. T. O. (1975a) A theoretical analysis of the contribution of algal cells to the attenuation of light within natural waters. I. General treatment of suspensions of pigmented cells. New Phytol., 75, 11-20. Kirk, J. T. O. (1975b) A theoretical analysis of the contribution of algal cells to the attenuation of light within natural waters. II. Spherical cells. New Phytol., 75, 21-36. Kirk, J. T. O. (1994) Light and Photosynthesis in Aquatic Ecosystems. Cambridge University Press, Cambridge, 2nd edn. Madden, C. J. and Kemp, W. M. (1996) Ecosystem model of an estuarine submersed plant community: calibration and simulation of eutrophication responses. Estuaries, 19, 457-474. Murray, A. and Parlsow, J. S. (1999) Modelling the nutrient impacts in Port Phillip Bay a semi enclosed marine Australian ecosystem. Mar. Freshwater Res., 50, 469-81. Murray, A. and Parslow, J. (1997) Port Phillip Bay Integrated Model: Final report. Tech. rep., CSIRO, GPO Box 1666, Canberra, ACT 2601. Nixon, S. W. (1995) Coastal marine eutrophication - a definition, social causes and future concerns. Ophelia, 41, 199-219. Pasciak, W. J. and Gavis, J. (1975) Transport limited nutrient uptake rates in Dictylum brightwellii. Limnol. Oceanogr., 20, 604-617. Riber, H. H. and Wetzel, R. G. (1987) Boundary-layer and internal diffusion effects on phosphorus fluxes in lake periphyton. Limnol. Oceanogr., 32, 1181-1194.

33

Technical Description of the CSIRO SERM Ecological Model

Sanford, L. P. and Crawford, S. M. (2000) Mass transfer versus kinetic control of uptake across solid-water boundaries. Limnol. Oceanogr., 45, 1180-1186. Shimeta, J., Jumars, P. A. and Lessard, E. J. (1995) Influences of turbulence on suspension feeding by planktonic protozoa: experiments in laminar shear fields. Limnol. Oceanogr., 40, 845859. Smith, F. A. and Walker, N. A. (1980) Photosynthesis by aquatic plants: effects of unstirred layers in relation to assimilation of CO2 and HCO- from cell volume or plasma volume. Limnol. Oceanogr., 12, 411-418. Thomas, F. I. M., Cornelisen, C. D. and Zande, J. M. (2000) Effects of water velocity and canopy morphology on ammonium uptake by seagrass communities. Ecology , 81, 2704-2713. Yool, A. (1997) The Dynamics of Open-Ocean Plankton Ecosystem Models. Ph.D. thesis, Dept. of Biological Sciences, University of Warwick.

34

This document was created with Win2PDF available at http://www.daneprairie.com. The unregistered version of Win2PDF is for evaluation or non-commercial use only.