Stevenson et al.: Patterns of species richness, diversity, population density, and distribution in the skates of Alaska 
27 
depth strata were combined into three depth ranges 
(<100 m, 101-200 m, and >200 m). 
In all surveys, skate specimens were either examined 
and discarded at sea or fixed onboard in 10% seawa- 
ter-buffered formalin or frozen for later study. Many 
of the specimens were photographed at sea, and the 
majority of fixed specimens were transferred to 70% 
ethanol and archived at the University of Washington 
(UW) fish collection. In compiling a list of skates for 
Alaska, we followed the comprehensive guide of Meck- 
lenburg et al. (2002), with the following updates. We 
included Rathyraja mariposa (butterfly skate), described 
in Stevenson et al. (2004), and Amblyraja badia (rough- 
shoulder skate), recently documented in the Bering Sea 
(Stevenson and Orr, 2005). 
Species richness was defined as the number of skate 
species encountered and identified in each haul. Mean 
species richness values were calculated for each depth- 
and-subregion stratum as a simple average of the num- 
ber of skate species encountered in all hauls performed 
in that stratum, including zero values for hauls in which 
no skates were encountered. Because mean species rich- 
ness figures were heavily influenced by zero values in 
some regions, and because simple species richness gives 
no indication of evenness, we also calculated a single 
Shannon’s diversity index ( H ' ) for each depth-and-sub- 
region stratum. This index was calculated as 
s 
H' = ^[(re- / re)ln(re ; - / re)], 
i=l 
where n i = the number of individuals belonging to the 
zth of S species in the depth-and-subregion 
stratum; and 
n - the total number of individuals captured in 
the depth-and-sub-region stratum (Ludwig 
and Reynolds, 1988). 
Bottom area swept was calculated for each haul by 
multiplying the distance fished by the mean net spread. 
Density was calculated as the number of individuals 
per km 2 of bottom area swept. Means are presented 
± standard error (SE). Because the distributions of 
both species richness and density values were heavily 
skewed, and had a large proportion of zero values, tests 
for significant differences among means were not per- 
formed. Species distributions represent a summary of 
all recorded encounters of each species during standard 
survey operations in successful survey hauls. 
Results 
The total amount of effort represented in this data set 
consisted of 6096 successful bottom trawls covering over 
201 km 2 of bottom area (Table 1). The Bering Sea shelf 
accounts for significantly more trawling effort than any 
of the other regions, because of differences in survey 
frequency (annual [Bering Sea shelf] vs. biennial [other 
regions]) and standard tow duration (30 minutes [Bering 
Boundaries of sampled subregions covered in 
bottom-trawl surveys from 1999 through 2004 
in (A) the eastern Bering Sea shelf (underlined 
numbers) and slope (numbers not underlined), 
( B ) the Aleutian Islands, and (C) the Gulf of 
Alaska. 
Sea shelf and slope] vs. 15 minutes [ Aleutian Islands and 
Gulf of Alaska]). Thirteen of the 14 described species of 
skates known from Alaskan waters (all but Rathyraja 
violacea) were encountered in bottom-trawl surveys from 
1999 through 2004. Survey trawls collected 0-7 species 
per haul (mean=0.93 ±0.01), and at least one species of 
skate was collected in 62.8% of hauls. The number of 
