currently no such index that has been developed for application in shelf waters 
throughout the entire West Coast. 
In the absence of a benthic index, we have attempted to assess potential 
stressor impacts in the present study by looking for obvious linkages between reduced 
values of key biological attributes (numbers of taxa, diversity, and abundance) and 
synoptically measured indicators of poor sediment or water quality. Benthic attributes in 
these offshore shelf waters showed significant variations among the three states. Thus, 
low values of species richness, H', and density were defined for the purpose of the 
present analysis as the lower 10 th percentile of values within each individual state. 
Thresholds for assessing poor sediment or water quality were defined as follows (sensu 
EPA 2000a for dissolved oxygen, EPA 2004 for other indicators): > 5 chemicals in 
excess of ERLs (from Long et al. 1995), > 1 chemical in excess of ERMs (from Long et 
al. 1995), TOC > 5%, and DO in near-bottom water < 2.3 mg/L. Appendix Table 4 
provides a summary by station of each of these variables and flags those falling within 
the defined levels of concern. 
This analysis revealed no major evidence of impaired benthic condition linked to 
measured stressors. There were only two stations, both in California, where low values 
of any of the three benthic attributes co-occurred with high sediment contamination or 
low DO in bottom water. One station (CA03-4039 off Los Angeles) had low benthic 
species richness and abundance accompanied by high sediment contamination, with 
eight chemicals in excess of corresponding ERL values and two in excess of ERM 
values. The other station (CA03-0059 north of San Francisco Bay) had low species 
richness and diversity accompanied by low DO. There were five other stations with DO 
in bottom water < 2.3 mg/L; however, none of these had low values of the three benthic 
variables. There were two stations (CA03-4030, CA03-4417) that had TOC levels in a 
range (> 5%) potentially harmful to benthic fauna. A third station (CA03-4430) showed 
a potential concern level if the more conservative threshold of 3.6% TOC is used 
(Hyland et al. 2005), but low values of benthic community attributes were not observed 
at any of these sites. High sediment contamination was a more prevalent stressor, 
occurring at 23 stations (all in California), but not at any of the sites where low values of 
benthic attributes were observed. In fact, most of these latter stations with high 
sediment contamination had more than 100 species grab' 1 . 
Such lack of concordance suggests that these offshore waters are currently in 
good condition, with the lower-end values of the various biological attributes 
representing parts of a normal reference range controlled by natural factors. Multiple 
linear regression was performed using full model procedures to test for the significance 
and direction of relationships between each of the benthic variables and various abiotic 
environmental factors (latitude, depth, percent fines). Data transformations were made 
where needed (i.e., square root for richness, logio for abundance) to meet analysis 
assumptions including normality and homoscedasticity of residuals. Results (graphics 
not shown) suggested that latitude and depth had significant influences on benthic 
variables region-wide. All three benthic variables showed significant inverse 
109 
