the microcosm and experimental ecosystem (i.e., natural ecosystem) was observed.
The lower limit of the confidence interval is a straight line obtained by dividing the
average value of the NOEC of the natural ecosystem by the uncertainty factor
(coefficient of assessment ¼ 200), obtained by considering the difference in ecosystem
sensitivity. Only the lower limit is needed to predict the NOEC in the natural ecosystem. If the correlation coefficient is greater than this lower limit (i.e., plots above the
lower limit line), the concentration obtained from dividing the NOEC measured using
the microcosm by the coefficient of assessment (i.e., the predicted NOEC) will be
lower than the predicted NOEC (PNOEC) of the natural ecosystem (i.e., the mesocosm
experiment); this means that it can predict natural ecosystems. In this study, most of the
measured substances were contained in this numerical region (i.e., above the lower
limit line). However, although substances exceeding the upper limit can predict the
PNOEC, the value is smaller than the NOEC of the natural ecosystem. Only one
substance out of 26 substances exceeding the upper limit could be accurately predicted
using the microcosm test, regardless of the substance. Additionally, a comparison of
the PNOEC of the natural ecosystem and the microcosm NOEC, as well as various
techniques, are shown in Fig. 8.3. The microcosm test has equivalent PNOEC and
NOEC, there is little variation, and precision is high. Moreover, an impact statement is
possible for ecosystemic sensitivity at approximately the same level, in comparison
with certain ECETOC and HC5 methods that are more conventionally used.
8.2 Comparison and the Interface of the Microcosm
and Natural Ecosystem (Mesocosm)
When the validity of the microcosm test was considered, it was shown that
23 materials fell within the confidence interval of the regression line out of
26 total materials, and the strength of the correlation between the experimental
ecosystem (microcosm) and the natural ecosystem (mesocosm) was illustrated by
the correlations between the means of the NOEC of the microcosm N-system and
the NOEC of the natural ecosystem (Fig. 8.4). Most materials that were measured
in this study fit into this domain (i.e., above the lower limit), as described in Sect.
8.1. However, for the material value that was greater than the upper limit, the value
was less than that of the natural ecosystem NOEC, although it could be used to
predict the PNOEC. The characteristics of the chemical substance used in the
microcosm test are shown in Table 8.1.
Two substances that fall outside of the confidence interval of the correlation
between the microcosm NOEC and the average NOEC of the natural ecosystem
⁄
ä
Fig. 8.1 (continued) test and the NOAEC in the outdoor ecosystem test (i.e., the mesocosm test) is
equal to 0.57, the range over which only σ moved the regression line downward may account for
84% of the chemical substance
8 Coefficient of Assessment for the Microcosm System
161
The lower limit of the confidence interval is a straight line obtained by dividing the
average value of the NOEC of the natural ecosystem by the uncertainty factor
(coefficient of assessment ¼ 200), obtained by considering the difference in ecosystem
sensitivity. Only the lower limit is needed to predict the NOEC in the natural ecosystem. If the correlation coefficient is greater than this lower limit (i.e., plots above the
lower limit line), the concentration obtained from dividing the NOEC measured using
the microcosm by the coefficient of assessment (i.e., the predicted NOEC) will be
lower than the predicted NOEC (PNOEC) of the natural ecosystem (i.e., the mesocosm
experiment); this means that it can predict natural ecosystems. In this study, most of the
measured substances were contained in this numerical region (i.e., above the lower
limit line). However, although substances exceeding the upper limit can predict the
PNOEC, the value is smaller than the NOEC of the natural ecosystem. Only one
substance out of 26 substances exceeding the upper limit could be accurately predicted
using the microcosm test, regardless of the substance. Additionally, a comparison of
the PNOEC of the natural ecosystem and the microcosm NOEC, as well as various
techniques, are shown in Fig. 8.3. The microcosm test has equivalent PNOEC and
NOEC, there is little variation, and precision is high. Moreover, an impact statement is
possible for ecosystemic sensitivity at approximately the same level, in comparison
with certain ECETOC and HC5 methods that are more conventionally used.
8.2 Comparison and the Interface of the Microcosm
and Natural Ecosystem (Mesocosm)
When the validity of the microcosm test was considered, it was shown that
23 materials fell within the confidence interval of the regression line out of
26 total materials, and the strength of the correlation between the experimental
ecosystem (microcosm) and the natural ecosystem (mesocosm) was illustrated by
the correlations between the means of the NOEC of the microcosm N-system and
the NOEC of the natural ecosystem (Fig. 8.4). Most materials that were measured
in this study fit into this domain (i.e., above the lower limit), as described in Sect.
8.1. However, for the material value that was greater than the upper limit, the value
was less than that of the natural ecosystem NOEC, although it could be used to
predict the PNOEC. The characteristics of the chemical substance used in the
microcosm test are shown in Table 8.1.
Two substances that fall outside of the confidence interval of the correlation
between the microcosm NOEC and the average NOEC of the natural ecosystem
⁄
ä
Fig. 8.1 (continued) test and the NOAEC in the outdoor ecosystem test (i.e., the mesocosm test) is
equal to 0.57, the range over which only σ moved the regression line downward may account for
84% of the chemical substance
8 Coefficient of Assessment for the Microcosm System
161