preseatat
Por eech of the ten primary subjects of investigation
(survey meter readings, 9011, veter, plankton, algae, land
pleats, invertebrates, fish, birds, and rets), the trends or
Geclinse are shown grephicelly, and in some cases also in
tabular forms.
Por a1] subjects the regressions along vith
Peleveat data are brought together in Table 1. Where avatiable
the pre-Bectar level appears near the left edge of the dec] ine
ereph as either a short horizontal] bar or vedge.
Por the straight lines depicting the declines where linearity appears to prevail, the time span involved is stipulated
in Table 1 as well as being shown by the absciseal range of the
lines in the grephs.
Por conversion between microcuries and disintegretions per
minute the following relationship was employed:
1 uc » 2.2 x 10° d/a.
The log-log regression line is determined by its slope and
y-intercept on day number 1, according to the relationship:
Ys at?,
where Y is the amount of radioactivity at time t in days after
assumed detonation day, and a is the y-intercept expressed in
units of radioactivity of the regression line of slope b on day
number 1. Por example, the second entry in Table 1, survey
meter read'ngs at Belle, graphed in Figure 3, involved observa-
tions on 16 days over the period 5-540 days after Nectar.
regression vas
.
The
Ye 2.5 x 103 t72-)* apr,
with a correlation of -.971, which is far beyond the 1% level
of P.
Along vith decline data, available decays for as nearly
simultaneous periods as possible are presented for ccmparison.
Decays start later than declines because declines vere corrected
back to date of collection, while decays are for the actual
dates of counting.
On the decay graphs the ordinate represents gross beta plus
the negligible alpha and ganma activity that would be detected.
Decay curves even on the same graph are not comparable to
one another as to absolute levels, because of vertical shifting
to obtain compact presentation, but may be compared as to slope.