We used the 3-D SYMPHONIE ocean circulation model (Marsaleix et al., 2009a, b, 2012). This model has been widely used in the Mediterranean Sea to study different marine processes related to coastal circulation (Estournel et al., 2003; Petrenko et al., 2008), sediment transport (Ulses et al., 2008), larval dispersal (Guizien et al., 2012) and plankton population dynamics (Auger et al., 2011; Herrmann et al., 2014). This model has also been used, for the first time, in the western North Pacific Ocean to study the 137Cs dispersion after the FNPP accident (Estournel et al., 2012).

The numerical configuration used in this study was the same as the one reported in detail by Estournel et al. (2012), with 30 vertical irregular levels based on the sigma coordinate system and characterized by an increase of resolution near the surface. The horizontal grid (Fig. 1) corresponds to an orthogonal curvilinear system, with variable resolution increasing linearly with the distance from FNPP (0.6 × 0.6 km near FNPP and 5 × 5 km at the open lateral boundaries off Japan).

To properly represent the dynamics of the plankton populations exposed to the radioactive contamination in our study area, the NEMURO (North-Pacific Ecosystem Model for Understanding Regional Oceanography) biogeochemical model (Kishi et al., 2007) was applied. This model, which has been extensively used in the western North Pacific region (Aita et al., 2003; Hashioka and Yamanaka, 2007; Komatsu et al., 2007), consists of 11 state variables with two size classes of phytoplankton: small phytoplankton (PS) representing small species such as coccolithophorids and flagellates, and large phytoplankton (PL) representing diatoms. It includes three size classes of zooplankton: small zooplankton (ZS) such as ciliates and foraminifera, large zooplankton (ZL) (copepods) and predatory zooplankton (ZP) such as krill and/or jellyfish. The other model state variables are: nitrate (NO3), ammonium (NH4), silicate (Si(OH)4), particulate organic nitrogen (PON), biogenic silica (Opal) and dissolved organic nitrogen (DON). The model structure and the different parameter values are presented in detail in Kishi et al. (2007).

The ocean circulation model (OCM) was run from February 2010 to January 2013. The currents, vertical diffusivities and temperature fields were then used to force the ecosystem model and spun up for 3 years by repeating the same forcing data for the first 2 years. For this study, we used the results of the 2 last simulated years (February 2011 to December 2012), when a quasi-steady state was reached.

To assess the effect of the accident on the planktonic populations, two different simulations were carried out: (1) the real (accidental) situation with presence of contaminated waters due to the accident that occurred on 11 March 2011, and (2) a non-accidental situation by assuming homogeneous 137Cs concentration in seawater over the whole simulation period.

Before the accident date (11 March 2011), the seawater 137Cs concentration for the western North Pacific Ocean ranged from 1 to 2 mBq L-1 (Povinec et al., 2013). For the purposes of the modeling, a constant 137Cs concentration in seawater of 2 mBq L-1 is assumed throughout the study area. In the accidental situation, we used, as of 11 March 2011, the 137Cs concentrations in seawater obtained from the dispersion simulation carried out by Estournel et al. (2012), in which the amount of atmospheric deposition included was 0.26 PBq within a radius of 80 km. The direct leakage was about 5.5 PBq released between 12 March and 30 June 2011. The simulation was extended until 31 December 2012, and the inverse method described in Estournel et al. (2012) and used to calculate the source term in the first three months after the accident was applied to the whole period. After June 2011, the concentrations at the two outlets of the nuclear power plant were simplified to a linear decrease from 40 and 20 Bq L-1 on 1 July 2011 to 8 Bq L-1 for both outlets at the end of 2011 and then remained constant at this value for 2012. However, no other additional source (e.g., terrestrial runoff, rivers flow, etc.) has been considered in this simulation.

The radioecological parameters related to plankton are very scarce, and are often associated with considerable uncertainties. In this study, a temporal series of the 137Cs concentration in zooplankton collected at Sendai Bay between June 2011 and December 2013 and reported in Kaeriyama et al. (2014) was used to calibrate the model and estimate the different radioecological parameters. However, because of non-indication of the zooplankton taxa composition, we used for the purpose of modeling a weighted average of 137Cs concentrations in the three zooplankton groups.

To assess the sensitivity of the calibrated parameters, we investigated a sensitivity analysis of the radioecological model using the classical one-parameter-at-a-time analysis (OAT). The choice of this quantitative method can be justified by its simplicity and by the absence of any interactive effects among parameters. In this local approach, the single-parameter variation effect is estimated by increasing and decreasing each parameter in Eqs. (3) and (7) by 10 %, while keeping all the others fixed at their nominal values. The sensitivity Sp associated with each parameter p was computed as the percentage of change in activity generated by the parameter variation: Sp%=Ep-EE×100, where Ep is the prognostic variable value (here, the 137Cs concentration in plankton populations) when the parameter p is set to its changed value (10 % higher or lower than its calibrated value), and E is the value of the prognostic variable in the baseline run (i.e., all parameters at their calibrated values).

To assess the biological effects of the 137Cs ionizing radiation on the plankton populations, we calculated the absorbed dose rate from internal and external pathways using the ERICA graded approach (Brown et al., 2008). This approach consists in converting the 137Cs concentration in plankton populations and in seawater to the internal and external absorbed dose rates, respectively, using the so-called “Dose Conversion Coefficients”, which are specific for each radionuclide–organism combination. The different dose rates are calculated as follows, assuming that the organisms are freely floating in the water column without any contact with sediment: D=Dint+Dext Dint=DCCCs-pk[Cs]pk Dext=DCCCs-w-pk[Cs]w, where D, Dint, Dext are, respectively, the total, the internal and the external dose rates (µGy h-1), the [Cs]pk and [Cs]w are, respectively, the 137Cs concentration in plankton population and seawater (in Bq kg-1),the DCCCs-pk is the dose conversion coefficient for the internal exposure, and DCCCs-w-pk represents the dose conversion coefficient for external exposure (in µGy h-1 per Bq kg-1).

The DCC parameter values for phytoplankton and zooplankton used in this study are obtained from the coastal aquatic ecosystem DCCs reported in Vives i Batlle et al. (2004). The values of these parameters are summarized in Table 1.

The seasonal variations in phytoplankton and zooplankton biomasses were presented for three different areas classified according to latitude: the subtropical region (latitude < 35∘ N), the transition region (35∘ N < latitude < 39∘ N), and the subarctic region (latitude > 39∘ N) (Fig. 1). The ecosystem model outputs are expressed in µmol N L-1, their conversion to the chlorophyll-a unit is carried out using a typical C : chlorophyll ratio of 50, and a C : N ratio of 133/17 (Kishi et al., 2007).

The monthly medians of the spatial chlorophyll-a concentration averaged over a 50 m deep layer were used to compare model results for the period (2011–2012) with the 20 years of climatology field data (1990–2010) (Fig. 2a, c, e) derived from the Japan Oceanographic Data Center (JODC) data set (available at: http://www.jodc.go.jp). In all areas, the temporal evolution of the chlorophyll standing stocks showed a seasonal cycle with higher median values in spring (April–May) and autumn (October–November). This seasonal cycle is less marked in the subtropical region than in the two other regions. The simulated chlorophyll-a concentration medians varied from less than 0.5 mg m-3 in all regions in winter to approximately 1, 1.5 and 3 mg m-3 in spring in the subtropical, the transition and the subarctic regions, respectively. These values of the chlorophyll-a concentrations are in general consistent with the field data, and show the same seasonal variability (Wilcoxon rank sum test (α=0.05): P=0.88).

The total zooplankton biomass and its taxonomic composition are presented in Fig. 2b, d, f for the three regional areas described above. The simulated zooplankton biomasses showed an annual seasonality in the three regions, with minimum values in winter and peaks in spring and autumn. The zooplankton biomasses showed latitudinal variations with greater biomass in the subarctic region (from 200 mg m-3 wet weight in winter to about 700 mg m-3 wet weight in late spring), followed by the transition region (from 150 to about 500 mg m-3 ww) and the subtropical region (from 100 to about 300 mg m-3 ww).

In the subtropical region, the taxonomic composition of zooplankton biomass was dominated by large zooplankton with about 40 %, followed by small and predatory zooplankton each accounting for 30 % of the total biomass.

In the transition region, the seasonal cycle of zooplankton composition was more pronounced. In winter, the zooplankton was represented by 40 % of large zooplankton, and 30 % of small and predatory zooplankton. In spring, the zooplankton biomass was dominated by large zooplankton (60 % ZL and 20 % for both ZS and ZP). From late spring until early autumn, the zooplankton composition changed progressively with a decrease of the ZL proportion, to be composed of 40 % ZP and 30 % of ZS and ZL in early autumn.

In the subarctic region, the proportions of small zooplankton, large zooplankton and predatory zooplankton were, respectively, 25, 35 and 40 % in winter, 10, 70 and 20 % in spring, and 20, 35 and 45 % in late summer and early autumn.

The result of the calibration is shown in Fig. 3, and the final estimated radioecological parameters are summarized in Table 2. The phytoplankton elimination rates estimated from this calibration (0.5 d-1) were very similar to that calculated using the allometric relationship reported by Vives i Batlle et al. (2007) (0.58 d-1). For the zooplankton, the obtained values ranged from 0.03 to 0.11 d-1, and are also in good agreement with the literature values (Thomann (1981): 0.03 d-1; Vives i Batlle et al. (2007): 0.056 d-1).

The 137Cs assimilation efficiency by zooplankton calibrated in this study was 0.75. This value is similar to that used by Brown et al. (2006), and is slightly higher than the 0.63 observed by Mathews and Fisher (2008) for the crustacean zooplankton Artemia salina.

The rates of 137Cs direct accumulation from water by zooplankton found in this study were about 5 × 10-4 L g-1 for small and large zooplankton, and about 0.001 L g-1 d-1 for predatory zooplankton. The accumulation rate corresponding to phytoplankton was 0.015 for both groups.

However, for the calibration we used zooplankton data from coastal areas presented in Kaeriyama et al. (2014). According to these authors, zooplankton gut content in these areas may contain particles with high 137Cs levels, which could affect the calibrated values. Consequently, overestimations in 137Cs concentrations in these populations could be generated especially in the open ocean where the particles contribution is generally negligible.

The sensitivity of the estimated 137Cs activity concentrations in different plankton groups to uncertainty in the parameters of Eqs. (3) and (7) calibrated to field data at Sendai Bay was tested using the OAT method, and the results are shown in Fig. 4.

For all plankton groups, the 137Cs activity estimates showed a great sensitivity to the 137Cs concentration in seawater, with an activity change of 10 % for a 10 % change in the seawater 137Cs concentration. The 137Cs activity in seawater used in this study was obtained from the numerical simulations of the 137Cs dispersion using the SYMPHONIE circulation model. One can imagine that all potential biases associated with this simulation would generate the same ranges of error in the results concerning the 137Cs concentration in plankton. It is, therefore, clearly important to take into consideration all these errors when interpreting the results of the radioecological model.

The 137Cs activity estimates in the phytoplankton groups are very sensitive to the accumulation rate from water (10 % change for a 10 % change in the parameter), and are moderately sensitive to the elimination and primary production rates (5–7 % change in the opposite sense), whereas the sensitivity to the daily respiration rate did not exceed 1 %. The primary production rate is, therefore, the most important ecological parameter in the estimation of 137Cs concentrations in phytoplankton. It allows dilution of the 137Cs concentrations in phytoplankton by promoting the growth of its populations.

For all zooplankton groups, the activity estimates were most sensitive to the change in the 137Cs assimilation efficiency (AE), with an activity change of about 9 % for both small and large zooplankton. For predatory zooplankton, the activity change was slightly above 10 %, which can be explained by the direct effect of the AE parameter on ZP and the indirect effect due to the change in ZS and ZL that are preyed on by ZP.

The sensitivity to the population growth efficiency (b) was also significant with about 7 % of change. This ecological parameter, which affects the zooplankton population growth and consequently plays a role in the dilution of their 137Cs concentrations, is associated with substantial uncertainty. Sushchenya (1970) reported values ranging from 4.8 to 48.9 %. The value used in this study was 30 % (Kishi et al., 2007). One can expect, therefore, an overestimation of up to 45 % or an underestimation of up to 60 % in the estimates of zooplankton 137Cs concentrations.

The sensitivity to the direct accumulation rate of 137Cs from water by zooplankton (μz) was relatively low (< 4 % for the three groups of zooplankton). This can be related to the lower proportion of contamination coming from water compared to that coming from food. The variation in the depuration rate induced a relatively moderate change of 5 %.

The sensitivity of the 137Cs activity estimates in the three groups of zooplankton to parameters related to their different preys is also not negligible. The proportions of change varied from 1 to 9 % depending on the zooplankton group and the parameter in question. For example, the sensitivity of the 137Cs concentration in ZS to the PS accumulation rate (μps), the elimination rate (λps), and the primary production rate (P) were 9, 5 and 7 %, respectively.

This sensitivity analysis showed that the parameters related to the two groups of phytoplankton are very important for the estimation of the 137Cs concentration in all plankton groups. Therefore, these parameters are key determinants of the radionuclide concentration in all marine animals of the pelagic food chain (Mathews and Fisher, 2008). Consequently, the experimental determination of these parameters, often neglected due to the difficulties characterizing the measurement of radionuclides in phytoplankton, is of the greatest importance.

The simulation results corresponding to the spatial distribution of the weighted average of 137Cs concentrations in the three zooplankton groups (ZS, ZL, ZP) are presented in Fig. 5. These results are shown for six different dates from June 2011 to August 2012, and are compared to the few field observations available in the area (Buesseler et al., 2012; Kaeriyama et al., 2014; Kitamura et al., 2013). Some field data reported in the unit of Bq kg-1 dry weight are converted to Bq kg-1 wet weight using dry to wet weight ratio of 0.2 (Buesseler et al., 2012).

In general, these results illustrated the good agreement between measured and simulated results, which is confirmed by the Wilcoxon rank-sum statistical test (P>0.05, non-significant difference). Nevertheless, some points showed significant discrepancies between measured and simulated concentrations, as in the case of (36∘ N, 144∘ W) in the period 3–18 June 2011 (Fig. 5a) where the observed concentration was 2 orders of magnitude higher than the simulated one. A large part of this difference could be due to a spatial shift of the contaminated plume in the dispersion model. Indeed, the coastal waters off Japan are very energetic, especially with the interaction between the cold Oyashio current moving southward and the warm Northward Kuroshio current, generating very complex physical structures (eddies, tidal forces, etc), which are generally less well represented by the hydrodynamics models leading to some spatial shifts between the simulated 137Cs concentrations in seawater used in this simulation and the real field concentrations.

To assess the contamination level of plankton populations in 2011, we calculated a ratio (R) of the137Cs concentration in phytoplankton (the weighted average of PS and PL) and zooplankton (the weighted average of ZS, ZL and ZP) in the accidental situation to its concentration in these populations in the non-accidental situation. The results of the temporal evolution of these ratios for different distances from FNPP are shown in Fig. 6.

The ratios for phytoplankton and zooplankton are very similar spatially and temporally. After the accident, the ratio increased rapidly until reaching a maximum, whose value and the time required to reach it are variable following the distance from FNPP. The results showed that the time, calculated from the accident date, required to reach the maximum value increased with distance from FNPP, going from about 1 month for the populations located at less than 30 km from FNPP to about 6 months for those located at 500 m from FNPP. The maximum value, in turn, decreased with the distance from FNPP (about 104 at 0–30 km from FNPP to slightly lower than 102 at 400–500 km from FNPP).

After reaching the peak, the ratios progressively decreased over time but remained relatively high at the end of 2011 especially in the sectors situated at less than 50 km from FNPP where the ratio was still higher than 10.

The rapid decrease of 137Cs in planktonic populations 1 year after the accident in the major parts of the study area can be explained by the different processes related to both population ecological functioning (cells growth and death, biological elimination) and the surrounding environment conditions, especially by the horizontal and vertical mixing due to the ocean hydrodynamics. Indeed, FNPP is located in an area where the east-flowing Kuroshio current and the southwest-flowing Oyashio current mix, generating complicated nearshore currents and mesoscale eddies (Buesseler, 2014), thereby favoring dispersion, regeneration, and thus dilution, of the contaminated planktonic populations in the area.

Referring to the biogeochemical cycle in the pelagic environment, part of the contaminated populations would be transferred to the pelagic higher trophic levels (planktivorous fishes, squids, etc.) by predation, leading to transfer of this contamination along various trophic chains. The other part will generate, after dying, large aggregated particles, known collectively as marine snow, which can reach the deep waters (Asper et al., 1992) and thus contribute to the contamination of sediment and benthic organisms, especially in the coastal area. This phenomenon was observed in the Mediterranean Sea a few days after the Chernobyl accident, generating a rapid transport of some radionuclides from surface waters to a depth of 200 m (Fowler et al., 1987). This process could be expected in the Japanese coastal area characterized by very high levels of contamination, especially around FNPP.

The concentration ratio (L kg-1) is defined as the ratio of radionuclide in the organism (Bq kg-1 wet weight) divided by its concentration in the water (Bq L-1). The dynamics of the calculated apparent concentration ratios (aCR) for small phytoplankton, small zooplankton and predatory zooplankton populations throughout the study area and for populations located within a radius of 30 km from FNPP over the year 2011 are shown in Fig. 7. These apparent concentration ratios are estimated for the two different situations described above (see Sect. 2.4).

The spatial median of the apparent concentration ratios in the non-accidental situation (i.e., the steady-state situation) was between 20 and 30 L kg-1 wet weight for small phytoplankton and between 10 in winter to slightly more than 30 L kg-1 during the rest of the year for small zooplankton. In the case of predatory zooplankton, the concentration ratio was a little higher, ranging from 10 to about 40 L kg-1 wet weight. These values are in good agreement with the reported data on plankton concentration ratios in marine ecosystems, which generally range from 6 to 40 L kg-1 wet weight in steady-state conditions (Fowler, 1977; IAEA, 2004; Kaeriyama et al., 2008). In the sector situated at less than 30 km from FNPP (Fig. 7), the concentration ratio was almost constant and seasonal variability was very less pronounced, with about 25 L kg-1 for PS and 30–40 for ZS and ZP. This constancy in the estimated concentration ratios for the populations located at less than 30 km compared to those estimated for the whole study area, where a substantial decrease in the concentration ratio was observed during winter, can be related to the clear differences in food ingestion rates observed in this period between the two locations (Fig. 8). In winter, the zooplankton ingestion rates estimated for the populations located at less than 30 km were higher than those estimated for the whole study area, due essentially to the spatial heterogeneity characterizing the whole study area in terms of food availability, with the presence of some less productive regions such as the subtropical zone where the planktonic biomasses were generally very low (see Sect. 3.1).

At the time of the releases and immediately after the accident, the concentration ratio decreased rapidly for all plankton groups. This is mainly due to the sudden arrival of highly contaminated waters in these areas where the living plankton populations were not yet contaminated. This phase was less marked for small phytoplankton compared to the groups of zooplankton, due to the fact that phytoplankton accumulates 137Cs only from water whereas in the case of zooplankton an important part of the contamination arises from food, a process requiring some time. For the populations located at less than 30 km from FNPP the dramatic decrease in the concentration ratio in March was even more intense and longer. The estimated time needed for these populations to regain the equilibrium was about 5–10 days for PS, 30 days for ZS and about 50 days for ZP. The decreasing phase in concentration ratio was directly followed by an increasing phase reflecting the progressive accumulation of 137Cs by plankton organisms.

The dynamics of 137Cs fraction accumulated from diet by zooplankton populations is estimated for both accidental and non-accidental situations and in the two spatial scales (Fig. 9). This fraction remained stable in the case of zooplankton living at less than 30 km from FNPP and represented more than 80 % in the case of ZS, 90 % in the case of ZL and 98 % in the case of ZP. These results indicated that the major part of accumulated 137Cs by these populations is coming from food, which is consistent with the research conducted by Baumann et al. (2015), who postulated that the dietary route could be largely responsible for the 137Cs bioaccumulated by the zooplankton collected off Japan 3 months after the accident.

The accident effect was only briefly apparent with a slight decrease of this proportion.

Conversely, the proportion estimated for zooplankton populations living in the whole area revealed a decline in winter, especially in the case of ZS for which this proportion decreased to 30 %. Because of the non-decrease in the 137Cs concentration in PS during this period (Fig. 10), the decrease in the relative accumulation by ZS from diet could be related to the decrease in the food ingestion rate (Fig. 8). No apparent effect of the accident on the 137Cs fraction accumulated from diet was observed at this large spatial scale.

The trophic transfer factor (TTF), defined as the ratio of radionuclide concentration in the predator to its concentration in prey, was calculated for each zooplankton group. The small zooplankton (ZS) has only one prey (small phytoplankton), therefore the TTF was calculated directly by dividing the 137Cs concentration in the ZS by its concentration in the PS. In the case of large and predatory zooplankton that have more than one prey (three for each one), we considered the weighted average of the 137Cs concentration in preys related to each zooplankton group.

Boxplots of predicted TTFs over 2011 for the three zooplankton groups in the accident and steady-state situations are shown in Fig. 11 for the two spatial scales described above.

The predicted TTF medians in the steady-state situation for ZS, ZL and ZP were, respectively, about 1.5, 1.7 and 1.2 in the sector 0–30 km from FNPP, and about 1.2, 1.45 and 1.1 in the whole study area. The TTF values calculated for the whole study area were slightly lower than those of the 0–30 km sector, reflecting the variability in ingestion rate and diet composition between the two spatial scales (Fig. 8). The lower values of ZP TTFs compared to the two other zooplankton groups may also be due to differences in their respective ingestion rate values. The correlation coefficient r between the modeled TTF related to each zooplankton group in the steady-state conditions and their corresponding ingestion rates showed a good correlation for the three groups of zooplankton and in both considered spatial scales (Table 3).

The predicted TTFs in the accident situation were similar to those predicted in the steady-state situation when considering the whole study area. This is due to the fact that, in the farthest sites from FNPP, where the contamination was not very high, the return to equilibrium occurred more rapidly, leading to TTFs similar to those observed before the accident although the concentrations in the predator and its preys were higher than during the pre-accident period. In the sector 0–30 km from FNPP, the predicted TTFs in the accidental situation were lower than those predicted in the steady-state situation (non-accidental situation). This is due to the persistence of the non-equilibrium state and the high 137Cs concentrations in seawater in this area, and to the fact that zooplankton accumulates 137Cs mainly from food leading to a delay in its contamination compared to its preys.

In turn, the correlation coefficients between predicted TTFs and ingestion rates in the accident situation showed a very slight decrease when considering the whole study area, and a considerable decrease when considering only the sector 0–30 km from FNPP. This means that the instability and the non-steady-state conditions characterizing the post-accident period had significant effects on this correlation.

Previous works suggested that radiocesium is the only trace element apart from Hg that may be potentially biomagnified along food chains (Harmelin-Vivien et al., 2012; Heldal et al., 2003; Zhao et al., 2001). In our study, the modeled TTFs were generally higher than the unity for all zooplankton groups, showing evidence of biomagnification potential at this trophic level. Mathews and Fisher (2008) reached the same general conclusion for the crustacean zooplankton Artemia salina feeding on phytoplankton, and reported that TTFs are directly related to the food ingestion rates, and that a consistent capacity for biomagnification exists when the food ingestion rate is high.

The estimation of the absorbed dose rate (µG h-1) is an essential step enabling media/biota activity concentrations to be interpreted in terms of potential effect (Beresford et al., 2007) .

The calculated dose rates received by phytoplankton and zooplankton populations located at less than 30 km from FNPP over 2011 are shown in Fig. 12. The external dose rate was about 7 times higher than the internal dose rate for phytoplankton, and about 5 times higher than the internal dose rate in the case of zooplankton, resulting in similarity between the total and the external dose rates. The total dose rates for phyto- and zooplankton were also very similar, whereas the internal dose was higher for zooplankton than for phytoplankton.

For both phyto- and zooplankton, in the steady-state conditions before the accident, the dose rates were about 10-6 µGy h-1. The maximum value was reached 1 month after the accident with about 0.05 µGy h-1. From this date, the dose rates decreased progressively to reach about 5 × 10-5 µGy h-1 at the end of 2011. The calculated internal dose rates for zooplankton in June 2011 were about 10-4 µGy h-1, and were, therefore, about 5 times greater than those reported by Fisher et al. (2013) for copepods and euphausiids collected 30–600 km off Japan. This difference is mainly due to the fact that in this study the dose rates were calculated for the populations located at 0–30 km from FNPP, where the activity level of 137Cswas higher.

The maximum dose rates calculated here were very low relative to the benchmark value corresponding to 10 µGy h-1 as suggested by the ERICA approach (Beresford et al., 2007), signifying that the 137Cs levels were too low to cause a measurable effect on these plankton populations. However, this conclusion concerns only 137Cs – we ignore whether the ionizing radiation doses due to the other radionuclides released in high quantities following the FNPP accident, such as short-lived nuclides 132Te, 131I and 90Sr, can generate any effect on these populations. Finally, it is important to note that this finding may be representative of the average conditions characterizing the area located up to 30 km from FNPP, however in close vicinity to the FNPP (e.g., FNPP Port), the planktonic populations could have been exposed to more intense and more persistent doses that could generate higher deleterious effects on these populations.