A simple approach to better estimates of Sr-90 concentrations in crops

Beresford, N.A.; Barnett, C.L. ORCID: https://orcid.org/0000-0001-9723-7247; Chaplow, J. ORCID: https://orcid.org/0000-0002-8058-8697; Guillen, J; Lofts, S. ORCID: https://orcid.org/0000-0002-3627-851X; Kashparov, V.. 2022 A simple approach to better estimates of Sr-90 concentrations in crops. [Lecture] In: 5th International Conference on Radioactivity & Environmental Radioactivity, Oslo, Norway, 4-9 September, 2022.

Full text not available from this repository.

Abstract/Summary

To predict radionuclide concentrations in crops most predictive models use equilibrium concentration ratios (the dry mass activity concentration in plant relative to the dry mass activity concentration in soil). However, for a given radionuclide-crop type combination, concentration ratios can be highly variable (variation over three orders of magnitude is common). A considerable proportion of this variation is due to soil properties. In the 1990’s/2000’s ‘process-based’ models using relatively readily available soil parameters (e.g. percentage clay, exchangeable potassium content) were developed to make predictions of radiocaesium concentrations in crops. Such models offer an approach which give predictions appropriate to site characteristics and can also be applied spatially to identify ‘at risk’ areas in the event of a nuclear accident. They can also be used to predict the effect of some countermeasures (i.e. K-fertiliser application). Whilst these models were developed for radiocaesium and have received recent renewed attention prompted by the Fukushima accident, there has been no development and testing of a similar approach for the other likely long-term contaminant following a nuclear accident, Sr-90. In this presentation, we will report on our recent studies to develop process-based modelling approaches to predict Sr-90 activity concentrations in crops. We have successfully developed two approaches to predicting Sr concentrations in crops using soil parameters. The first of these used model reduction to adapt an existing, well established, chemical speciation model. Whilst we will give an overview of this work, the focus of the paper will be on a simpler approach, which only requires information on soil and crop calcium concentrations. To test the developed models we conducted greenhouse studies growing crops in six soil types from the United Kingdom and Spain. The crops grown were grass, lettuce, courgette, potato, chard, radish and strawberry. As already noted, the simpler of the two approaches developed requires soil and plant calcium concentrations. If we assume that either soil calcium concentrations will be available from measurements at a given site or national soil databases then we only need to estimate calcium concentrations in crops of interest. To enable this we have compiled a database of approximately 1000 values collated from various worldwide databases. Strontium concentrations in the crops grown in the greenhouse study are relatively poorly predicted using published concentration ratio values (International Atomic Energy Agency Technical Report Series 472) (R2=0.01). Predictions are considerably better using the simple approach we have developed (R2>0.5). In addition to the soil-plant model development and testing, we will also present a summary of our recent evaluation of the ability of suggested ‘phylogenetic models’ to predict Sr (and Cs) concentrations in crops. There is scope to use phylogenetic and process-based models in combination. Acknowledgements: We thank other staff at UK CEH Lancaster who have contributed to this work. The studies discussed here were conducted as part of the CONFIDENCE project, which is part of the CONCERT EJP funded by the European Union's Horizon 2020 research and innovation programme (grant agreement No 662287).

Item Type:	Publication - Conference Item (Lecture)
UKCEH and CEH Sections/Science Areas:	Pollution (Science Area 2017-)
Date made live:	03 Feb 2023 16:50 +0 (UTC)
URI:	https://nora.nerc.ac.uk/id/eprint/533136

Actions (login required)

View Item

Document Downloads

More statistics for this item...