Values of parameters and state variables considered in the numerical simulations.

aOnly Prorocentrum triestinum was considered because this species contributed to 99% of the total dinoflagellate abundance in Thau Lagoon (S1 Appendix, section S1-2).bMaximal growth rate of P. triestinum.cAverage value of half saturation constants for nitrate uptake of dinoflagellate species presented in Table 3.7 of this author.dNitrogen cell quota of Prorocentrum micans.eParameters of Amoebophrya sp. infecting Karlodinium micrum.fAverage value of the mean doubling rates (d–1) of Leptocylindrus minimus, Leptocylindrus danicus, Cylindrotheca closterium and Thalassionema nitzschioides (the dominant diatoms species in Thau Lagoon; S1 Appendix, section S1-2).gHalf saturation constant for nitrate uptake of Pseudo-nitzschia sp. shown in Table 3.7 of this author.hAverage values of nitrogen cell quota of diatom species presented in Table 1 of these authors.iAverage value of nanophytoplankton growth rates (d–1) presented by these authors in their Fig 4 (only control experiments, with no nutrient addition, were considered).jValue for nanophytoplankton natural assemblages in Thau lagoon presented by these authors.kAverage values of nitrogen cell quota of nanoplankton species presented in Table 1 of these authors.lAverage values of Tiarina fusum feeding on Lingulodinium polyedrum and Scrippsiella trochoidea (values in ng C–1 were transformed to cells L–1 considering carbon content per cell) given by the authors.mBased on average values estimated from growth rates of Brachionus plicatilis, Brachionus rotundiformis and Brachionus sp. feeding on Tetraselmis suecica and Nannochloris atomus (prey concentrations presented in ng C–1 were converted to cells L–1 by considering the cellular carbon content of a cell with equivalent spherical diameter of 10 μm and the equation for carbon conversion given by [29]).Values of parameters and state variables considered in the numerical simulations.