STOCHASTIC SIMULATION OF RADIONUCLIDE UPTAKE
803
rabbits were analyzed.'*»'? The amount of food ingested daily, D, was
taken as 100 g/day on the basis of work by Currie’ in Utah and by
Arnold!® in Arizona. The absorption and thyroidal uptake of ingested
1317 F,-was taken as 0.16 on the basis of research by French’ with
jackrabbits during the summer in Idaho. The effective half-life of
317 in the jackrabbit thyroid was estimated to be 2.5 days according to
French’s findings.
The function of the model, as a deterministic expression of what
is really a probabilistic process, has already been analyzed.’ It is
now instructive to consider the variability of the parameters influ-
encing the relation between '*!I on vegetation and in the thyroids of
herbivores. From such a consideration we may develop a hypothesis
regarding the nature of the frequency distribution of '*'I in consumer
populations. The importance of such understanding has already been
discussed.
If the possible variation of a prediction for any time t is to be con-
sidered, the variability of each input parameter and the type of fre-
quency distribution generated by the interaction of these parameters
must be known, Although the variation of the inputs was known approximately, it was not possible to predict the interaction of the variables
theoretically. Hence the deterministic model was revised as a probabilistic simulation of the experience of an herbivore in an area con-
taminated by I, and the simulation was programmed for IBM-7094
computer solution.
The revision is a discrete version of the deterministic model,
with an interval of one day between meals. Whenthe interval is zero,
as if feeding were continuous, the discrete version is identical with
the continuous. The advantage of the discrete version is that it facili-
tates the simulation of the feeding experience of a group of individuals
over a period of time. The purpose of designing the stochastic version
of the model is to permit the variables to take on different values and
to examine the influence of chance variation in the inputs on the pre-
dictions of the model.
It is easier to visualize the stochastic model if one considers what
occurs when an herbivore occupying a fallout field consumes a Series
of meals, each containing a varying amount of '%!I. Ultimately one
wishes to estimate the amount of 1°, A, in the thyroid of the consumer
at the end of some arbitrarily defined period of time. For example,
the amount of thyroid ''I after 30 days is the sum of 30 terms. Each
term represents the '*!I remaining in the thyroid from one of 30 meals.
From a given meal containing ‘I, some fraction, F, is absorbed and
reaches the thyroid. The 30-day thyroidal uptake of the nuclide may
be given as follows:
F(D,C, + D.C, ... + DggC3)
(1)