Radioecological modeling of the 131I activity dynamics in different types of grass vegetation in the Chernobyl accident year*

The dynamics of 137Cs and 131I radioactivity in the crude biomass of the grass fodder and food vegetation in Mazovia, Poland, in 1986, the year of the Chernobyl accident, has been estimated. Density of 137Cs and 131I in the soil and vegetation have been measured as a function of rainfall and biomass density as of the time most of the fallout took place. A method is described to convert the instrumental data for the radionuclide activity dynamics in vegetation of one type to vegetation of other types. The results of such data conversion from lawn grass to other types of food and fodder grass vegetation are presented. A method is described for adjusting the dynamics of the radionuclide transport through the food chain components (pasture grass, green meat – milk – human body) by normalizing successively the estimated data in each next component for the average value of the instrumental data ratio to the estimated data in the preceding component. The proposed methods are intended to generate a mutually consistent base of estimated and reconstructed instrumental data: 137Cs and 131I activity in the atmosphere – rainfall – 137Cs fallout density on terrain – specific activity of 131I in vegetation. Such radioecological database will provide for a longer reliability of the estimated 131I specific activity dynamics in milk and in human body and, in the long run, when estimating the thyroid internal exposure doses.


Introduction
The paper deals with studying the dynamics of 137 Cs and 131 I radionuclides transport in the trophic chain: atmosp-here -soil -vegetation -dairy cow organism -milk -human body, following the accident at the Chernobyl NPP. In (Vlasov et al. 2019), the authors described the technology and generation of a mutually consistent in-put database for the computational model for the regions of Mazovia, Poland, and Bohemia, Czech Republic. The set of data in the base for the regions of interest included the major fallout period average or time dependencies of the 137 Cs specific volumetric activities in the atmosphere, the rainfall in the fallout period, and the 137 Cs fallout density. Homogeneous and heterogeneous cloud models were used for the mutual agreement of this dataset.
The materials of the Prague and Warsaw scenarios under the IAEA's EMRAS project (Krajewski et al. 2008; Bartuskova et al. 2009;Zvonova et al. 2010;IAEA-TECDOC-16782012 were used in (Vlasov et al. 2019) to generate such databases. The key conclusion made in (Vlasov et al. 2019) was that the use of mutually consistent data on the rainfall in the major fallout period, the specific volumetric activities of 137 Cs in the atmosphere, and its fallout densities was expected to lead to a major decrease in the uncertainties involved in the 137 Cs and 131 I transport in the food chain and, therefore, to more accurately reconstructed internal exposure doses to the populationresided in the contaminated areas. This statement was checked in (Vlasov et al. 2019b) dealing with the verification of a radioecological model using the instrumental data for the assessment of the dynamics of the 131 I activities in grass in Bohemia and Mazovia. With regard to the specific nature of the grass radiometric sampling point, extra studies were carried out in the territory of the Warszawa Obserw Astr weather station in Mazovia for the grass type identification (annual or perennial lawn grass). On the basis of the results analysis it was found that the dynamics of the estimated data for perennial lawn grass agrees better with the measurement data than for annual lawn grass.
The final objective in building radioecological simulation models is to study and assess and the regularities of internal radiation dose formation to the public after radiological accidents with the release of radioactive products into the environment. When simulation models are used, the best way to improve the reliability of the estimates is successive normalization of the data for the radionuclides activity in the given food chain component with allowance for the value of the average instrumental data ratio to the estimated data in the preceding element.
For instance, the residual errors of the estimated and instrumental data on the contamination of fodder and food vegetation with 131 I and 137 Cs can be used initially to adjust the calculation results for the dynamics of their specific activities in milk through the intake of radionuclides into the dairy cow body with green fodder. At the next stage, similar residual errors for food vegetation and milk can be also used to adjust estimated data of the radionuclide activities in the body of local residents through the dietary intake pathway. Such successive adjustment through the food chain is expected to lead to a major decrease in the uncertainties involved in the internal radiation dose assessments to the population resided in radioactively contaminated areas.
The most acceptable option for such successive adjustments is the amount of residual errors in the region of the maximum instrumental data values. The rationale for such selection is the direct proportion between the maximum values of the radionuclide activities in all series components in the atmosphere -vegetation -milk -human body food chain.
The Warsaw scenario includes instrumental data of the 131 I activities in lawn grass and milk. There is no instrumental data for the dairy cow fodder vegetation in the scenario, which makes it impossible to use the same normalization for milk.
The paper deals with the development of a method to convert actual instrumental data of specific radionuclide activities in vegetation of one type obtained at one measurement point to reconstructed instrumental data for vegetation of another type from other points, specifically, the grass radiometry data for the territory of a weather station in Warsaw to the pasture grass data from the milk radiometry points in the milk producing areas in Mazovia.
The purpose of the study is to develop and implement the method of converting instrumental data of the radionuclides activity dynamics in grass vegetation of one type obtained at one point to the activity of radionuclides in vegetation of other types at points with different 137 Cs fallout density and rainfall in the fallout period, as well as to develop the method for the consistent mutual agreement of estimated and instrumental data on the dynamics of the radionuclide transport in the atmosphere -fallout -fodder and food vegetation -milk -human body food chain.

Materials and methods
For analysis of the instrumental data on the parameters of the radiation situation in the central part of Mazovia in the year of the Chernobyl accident and radioecological modeling of the 131 I activity dynamics in different types of grass vegetation of the agroecological block of a radioecological simulation model was used (Vlasov 2013). This block is a system of linear differential equations describing in real time the dynamics of the atmospheric 131 I and 137 Cs radionuclide fallout onto the ground and vegetation, and the dynamics of specific activities for grass, wild and cultured vegetation, and fodder and food crops with regard for the growth of their biomasses. The biomass dynamics is calculated with account of the weather data for the accident year. The model's input data consists of the following set of dynamic parameters: specific volumetric activities and occurrence forms of radionuclides in the atmosphere, daily average air temperatures from the beginning of the vegetation period, rainfall in the period of and after the fallout deposition, and crop yields in the accident year. This dataset, along with the data on the 137 Cs fallout densities in residential areas (RA), is presented, to a greater or smaller extent, in the EMRAS project scenarios (Krajewski et al. 2008; Bartuskova et al. 2009; IAEA-TECDOC-1678 2012).

Results and discussion
The Warsaw scenario includes instrumental data of the 131 I specific activities only in lawn grass in the territory of the Warszawa Obserw Astr weather station in Warsaw.
The instrumental data of the 131 I specific activity in this grass has an abnormal outlier on the ninth day and the second activity peak on the 16 th day after the accident. A more detailed analysis shows the presence of two instrumental data series with practically identical rates of their exponential decrease (Fig. 1).
We shall note that the time dependencies for each series and the entire set of instrumental data have an exponential form. We took this peculiarity into account when comparing these with the estimated data.
There is no measurement data for the dynamics of the 131 I activities in food and fodder vegetation in the Warsaw scenario. Using a simulation model (Vlasov et al. 2019), however, it is possible to convert actual instrumental data for the lawn grass at the Warszawa Obserw Astr weather station to the reconstructed instrumental data for vegetation of other types in other RAs where milk radiometry was undertaken. In our case, the conversion was done for cultivated grasslands, the primary source of the radionuclides a intake with the dairy cow ration in Mazovia during and after the cow placement on summer feed. Such conversion can be done using the following ratios: direct conversion -conversion from instrumental data of the lawn grass measurements, 1 M grass Q to "measurement" data for vegetation of other types, M gr Q : The dynamics of the biomass density for vegetation of different types, specifically dairy cow fodder (pasture grass, sown annual grass for green fodder) and the human ration's vegetation component (annual and perennial green) was calculated based on the model in (Vlasov et al. 2019) using the Warsaw scenario data on the annual cycle of the daily average air temperatures, rainfall data, and the respective crop yields in the Chernobyl accident year. An example of the calculations is presented in Fig. 2. Table 1 presents data on the cumulative effective biological temperatures of the vegetation development phases and their respective occurrence times. It was assumed in the calculations that sown perennial grass was mown for being used as green fodder simultaneously after its ripening, lawn grass was mown when its biomass was 1 kg/ m 2 , and annual and perennial green was consumed as human food after its biomass gain phase was over. The biomass of the grass not mown was assumed to be equal to 0.2 kg/m 2 .
The results of such instrumental data reconstruction for different types of grass vegetation in the Brinow bis community (the maximum values have been recorded in the Warsaw milk producing area: the 137 Cs fallout density Figure 1. Instrumental data of the 131 I specific activity in lawn grass at the Warszawa Obserw Astr weather station: 1 -all data; 2 -exponential interpolation; 3 -series 1; 4 -interpolation 1; 5 -exponential regression 1; 6 -series 2; 7 -interpolation 2; 8 -exponential interpolation 2. Figure 2. Dynamics of the 131 I specific activity in grass vegetation (a) and estimated data of the 131 I specific activity in the atmosphere (b): 1 -perennial lawn grass, Warsaw Obserw Astr*; 2 -annual green, Warsaw Obserw Astr*; 3 -perennial green, Przasynski**; 4 -perennial lawn grass, County Ostrolecki**; 5 -cultivated pasture grass, County Ostrolekа**; 6-rainfall, meteo station Warsaw Obserw Astr* (* -direct data; ** -data with conversion for the minimum 137 Cs fallout deposition densities in a milk producing area).  I specific activity within the Brinow bis community in the Warsaw milk producing area: 1 -instrumental data; 2 -exponential interpolation; 3 -perennial lawn grass, direct calculation; 4 -cultivated pasture grass, reconstruction; 5 -annual greens, reconstruction; 6 -perennial greens, reconstruction; 7 -sown annual grass for green fodder, reconstruction; 8 -rainfall intensity. of 22.7 kBq/m 2 , and the rainfall of 7.3 mm in the major fallout period) from the perennial lawn grass at the Obserw Astr weather station with the rainfall of 0.8 mm and the 137 Cs fallout density of 3.3 kBq/m 2 , based on a direct calculation model, are presented in Fig. 3, and the crude biomass dynamics for the above grass vegetation types is presented in Figs 3, 4 also presents • direct instrumental data of the 131 I specific activity in grass specimens; • an approximation of the instrumental data using an exponential dependence; • the calculated 131 I activity in mown perennial lawn grass as of the measurement times.
According to the calculation data, the density of the cultivated pasture grass crude biomass as of 26 April 1986 was 0.55 kg/m 2 , this practically coincided with the Warsaw scenario data which said that it was the warm spring with the temperature of about 20 °C in Mazovia in 1986. The temperature conditions accelerated the growth of vegetation, especially the pasture grass ,its crude biomass density was about 0.4 kg/m 2 in late April, 1986. Figs 5, 6 present estimated dependencies of the 131 I specific activity in grass vegetation on its crude biomass and the rainfall for the fallout period, with regard to direct instrumental data of the 131 I activity dynamics and atmospheric occurrence forms, according to the Warsaw scenario.
The estimated data in Figs 5, 6 show that the specific activity of 131 I in grass decreases as its biomass grows (the growth is the faster, the larger the rainfall is), and increases as the rainfall grows (the growth is the faster, the lower the density of its grass biomass is). Of the same form is the data on the dependence of the maximum reconstructed instrumental data values for the 131 I activities in grass vegetation of different types on their biomass as of the fallout start time obtained based on the data of the 131 I specific activity dependencies (see Figs 1-4) and presented in Table 2.
The calculation results show that the mowing of lawn grass led to an increase in the growth rate of its net biomass. Owing to this, its activity decrease rate was much higher than for pasture grass, increasing with the growth in the grass cutting frequency. The activity of vegetation with no cutting decreased with time only due to the radioactive decay of 131 I, wind, the rainfall after the fallout was over, and the natural net biomass growth.

Key results
A method has been proposed and implemented to convert actual instrumental data of the radionuclide specific activity in vegetation of one type to reconstructed instrumental data for vegetation of other types.

Conclusion
The proposed and implemented method to convert instrumental data for the radionuclide activity in vegetation of one type to fodder and food vegetation of other types can be used to adjust the dynamics of the radionuclide human intake with the human ration's vegetation component, the activities of radionuclides in milk, and further in the chain of their intake with the dairy component.
The statistical parameters of the calculated data relations for the measurement time to the entire series of instrumental data can be used to estimate the uncertainties of the radioecological model as the whole.