498
GUSTAFSON, BRAR, AND MUNIAK
tribution from natural activity in the air. A known portion of eachfilter
was cut into small pieces and mixed with similar pieces from other fil-
ters to form monthly or semimonthly composites. These composites
were
petri
compressed into a disk and placed in a plastic petri dish. The
dish
was
then placed on top of a 4- by 4-in. Nal crystal for
gamma-ray analysis. Pulse-height analysis was made with a multichannel analyzer with IBM typewriter or punched-card readout. Both
Sample and detector were enclosed in 6-in.-thick steel shielding for
background reduction and stabilization. Counting times of 100 min were
sufficient to accumulate 100,000 or more net counts in the energy region from 0.1 to 2.0 Mev. Both the monthly and the semimonthly composites contained particulates collected from about 104 m’ of air. Suf-
ficient time had elapsed between the collection of the last filter and the
time of measurement to ensure that the radon and thoron daughter products had decayed to an insignificant intensity.
The concentration of the various gamma-emitting radionuclides
was determined by a computer solution of the requisite numberof simultaneous equations.’ During those times when fresh fission debris
was present, “ce, “4ce,gb, '8Ru, Ru, M's, *Zr—*Nb, and
140R4~—14°la were determined by this method. lodine-131 was not readily collected on these filters, and “Ba—'°La was used as an indicator
of the upper limit of ‘1 present in surface air at the time.! On rare
occasions *Te—182] and Mo have been observed in surface air; however, analyses of these nuclides were not included in the routine pro-
gram. In addition, “Mn and '4sb have been determined when their
presence in the gamma-ray spectrum was suspected. Successive spectra were taken of the same filter composite over a period of several
months. This has made it possible to obtain a more exact determination of the shorter lived components and hence has also improved the
estimates of the longer lived nuclides. Beryllium-7 has also been determined from the difference spectra obtained from successive measurements of air filters collected several months after the cessation of
testing. In the absence of fresh debris, ‘4Cce, ‘sb, "cs, ‘Ru, ®zr—
*SNb, and Mn were determined in the initial spectrum, and 'Be was
determined in the difference spectra.
During times of atmospheric testing of nuclear weapons and for
some time thereafter, the levels of gross activity in surface air are
relatively high. In addition, the gross activity undergoes seasonal variations, with maximum concentrations occurring in the spring and lower
values during autumn and winter. This seasonal variation is more pro-
nounced in the case of longer lived radionuclides, such as ‘°"Cs, as illustrated in Fig. 1. Spring maximums were observed in 1960 and 1961;
inthe absence of currentor recently concluded testing, a meteorological
origin for these maximums is strongly indicated. From 1954 through
1960 the maximum was observed at Argonne during April and May