( 
0) TO 
SB 
© 6 
E © 
© jt 
CL *- 
_ </) 
^ "o 
© © 
s © 
o o 
® X 
CL LU 
w © 
ro E 
© 3 
E o 
o © 
© < 
ro o 
c © 
8 o 
© 5 
o 
0 10 20 30 40 50 60 70 80 90 100 
0 10 20 30 40 50 60 70 80 90 100 
Percentage of Area/Volume Exceeding the Criteria 
Percentage of AreaA/olume Exceeding the Criteria 
Figure VI-10. Light area reflects amount of 'allowable' criteria exceedance defined as the area under the 
reference curve (light line). Dark area reflects the amount of 'non-allowable' criteria exceedance defined as 
the area between the attainment curve (black line) and the reference curve. 
being evaluated and refined is the 
Kolmogorov-Smimov (KS) test, which 
was originally developed to test for signif¬ 
icant differences between cumulative 
density functions (Haan 1977). The KS 
test is nonparametric and is based on the 
maximum difference between curves 
(Figure VI-11). The maximum difference 
is somewhat different than the area 
between the curves, which is the preferred 
indicator for assessing attainment. 
However, it can be shown that the 
maximum difference and the area 
between the curves are closely correlated 
and, therefore, evaluation of one will 
reflect an evaluation of the other. 
The KS test is well-documented and 
accepted in the statistical literature. Some 
refinements that may be necessary are 
currently being evaluated. Overall, 
however, the KS test has a strong potential 
for evaluating water quality criteria attain¬ 
ment in the Chesapeake Bay. 
Percentage of AreaA/olume Exceeding the Criteria 
Figure VI-11. Illustration of the basis of the Kolmogorov- 
Smirnov statistical test for identifying statistically significant 
differences between cumulative density functions. In this case, 
the test is applied to identify statistically significant differences 
between the reference and attainment curves. 
chapter vi • Recommended Implementation Procedures 
( 
