A subset of 1 3 sites from 1 991 was sampled every year to provide an estimate of between-year variation. An 

 analysis of variance was performed on the data from the between-year temporal replicates to test for a year 

 effect on the benthic index and a pairwise T-test was used to detect significant differences in benthic index 

 values paired by station between any two years. For both tests, the null hypothesis of "no significant difference" 

 was not rejected (p > 0.05). The benthic index values among years at the thirteen stations are compared 

 categorically in Figure 3-5. We defined a change in classification as a change from degraded to undegraded 

 (or wee versa). Benthic index values between 3 and 5 represented moderate or marginal conditions. A 

 change in classification from degraded to moderate or undegraded to moderate was not viewed as a 

 misclassification by the index. Although the range of values for any given station is sometimes large, the 

 classification of a station does not change from degraded to undegraded (or wee versa) except in the case of 

 stations 1 and 10. The classification of a station does change from degraded or undegraded to moderate for 

 many stations, however. This validation exercise showed that the benthic index maintained relative inter- 

 annual stability in addition to minimal within-year variability. 



X 



o 



•a 

 c 



c 



QQ 



12 



10 



8 



6 



4 



2 







-2 



-4 



-6 



1991 

 ■ 1992 

 ▲ 1993 

 D1994 



01 23456789 10 11 121314 



Station 



Figure 3-5. Comparison of benthic index values at stations that were sampled each year. 

 Horizontal lines at benthic index values of 3 and 5 indicate the boundaries of 

 degraded, moderate and undegraded benthic conditions. 



3-18 



