assumption should generally hold, though the assumption is likely to break down at 
large spatial scales such as comparing point samples to total assemblages (Lee et al., 
2003). 
The distribution of the extent of invasion based on the relative species richness 
of nonindigenous species for all 217 benthic samples is shown in Figure 3.3.3. Two 
thresholds seem intuitive in interpreting this metric. The first is simply that the site is 
“uninvaded” if there are no nonindigenous species. Across the West, nonindigenous 
species are absent in 15% of the samples. The second proposed threshold is samples 
in which nonindigenous species constitute >50% of the combined native and 
nonindigenous species. Since nonindigenous species constitute at least half of the 
classified taxa, these sites can be considered to constitute a non-native assemblage 
and are classified as “highly invaded”. Approximately 42% of the samples are classified 
as highly invaded based on this threshold. 
There appear to be substantial differences in the extent of invasion both 
geographically and by habitat type (Figure 3.3.4). To better highlight these differences, 
this analysis separates the San Francisco high marsh samples from the rest of the San 
Francisco habitats and Puget Sound from the rest of the Washington samples even 
though they were not originally identified as separate reporting units. To test for 
significance among locations, a Kruskal-Wallis one-way Analysis of Variance on ranks 
was performed on the values of %NIS Spp from California without San Francisco, San 
Francisco without high marsh, San Francisco high marsh, Oregon, coastal Washington, 
and Puget Sound. Based on this nonparametric test, there is a significant difference in 
the median values of %NIS Spp among these six geographical areas or habitat types (p < 
0.05). San Francisco habitats other than the high marsh were the most invaded with an 
average of almost 50% of the classified species per sample consisting of nonindigenous 
species. The high marsh in San Francisco was less invaded, but this pattern may at 
least partially reflect that these sites had relatively high proportions of oligochaetes and 
insects that were not identified to species. The other apparent pattern is that the 
intertidal benthos in Puget Sound is less invaded. On average, Puget Sound samples 
contained about 26% nonindigenous species compared to 40% to 44% for coastal 
Oregon and Washington. 
The extent of invasion can also be measured by the relative abundance of 
nonindigenous species. Using the same approach as with non-native species richness, 
the relative abundance of nonindigenous species is calculated as a percentage of the 
combined abundance of natives and nonindigenous species as: 
%NIS A bun = NIS A bun/(NIS A bun & Nat Abun ) *100 (Equation 3.3.2) 
where: 
%NIS Ab un = relative abundance of nonindigenous species per sample 
NIS A bun = abundance of nonindigenous species in sample normalized to 0.09 m 2 
Nat A bun = abundance of native species in sample normalized to 0.09 m 2 
35 
