Y. Facorellis et al., STUDY OF THE PARAMETERS AFFECTING THE CORRELATION OF BACKGROUND VERSUS COSMIC-RADIATION IN CO2 COUNTERS - RELIABILITY OF DATING RESULTS, Radiocarbon, 39(3), 1997, pp. 225-238
Systematic treatment of the data recorded by our guard counters and co
rrections introduced for meteorological factors has allowed observatio
ns on solar events clearly manifested in the readings. Examples are th
e solar flares of March 1989 and especially of June 1991, which caused
a ca. 10% decrease in the cosmic radiation flux reaching the counters
. A sinusoidal variation in the cosmic-ray flux with a period of one y
ear is also clearly manifested in the data. The observation that the b
ackground in the C-14 measurements depends on the intensity of the cos
mic radiation has led to the use of monthly correlations for the deter
mination of the best background value to be used in the age calculatio
ns. This reduces the error significantly. However, various factors suc
h as random statistical fluctuations of the background measurements ma
y affect the slope of the correlations and consequently the calculated
age of the samples. Long-term observations of the relation between ba
ckground values and coincidence counts have led to constraints in the
slope of the correlation. A simple extension of the fitting procedure
is explored, which maintains the physically meaningful range of the sl
opes, but is flexible to adjust for the seasonally varying contributio
ns to the variations of the cosmic-ray flux.