Radioactive C-14 is formed as a by-product of nuclear power generation and from the operation of nuclear fuel reprocessing plants like AREVA-NC La Hague (North France), which releases about 15 TBq per year of C-14 into the atmosphere. This article evaluates a recently improved radioecology model (TOCATTA-chi) to assess C-14 transfers to grassland ecosystems under normal operating conditions. The new version of the TOCATTA model (TOCATTA-chi) includes developments that were derived from PaSiM, a pasture model for simulating grassland carbon and radiocarbon cycling. The TOCATTA-chi model has been tested against observations of C-14 activity concentrations in grass samples collected monthly from six plots which are located around the periphery of the reprocessing plant. Simulated C-14 activities are consistent with observations on both intensively managed and poorly managed grasslands, but an adaptation of the mean turn-over time for C-14 within the plant is necessary in the model to account for different management practices. When atmospheric C-14 activity concentrations are directly inferred from observations, TOCATTA-chi performs better than TOCATTA (the root mean square error is decreased by 45%), but when atmospheric C-14 activity concentrations are not known and must be calculated, the uncertainty associated with the TOCATTA-chi model outcomes is estimated to be larger than the standard deviation of the observations. (C) 2013 Elsevier Ltd. All rights reserved.