Burger 



Chapter 29 



Marine Distribution, Abundance, Habitats in British Columbia 



Table 2 Mean densities of Marbled Murrelets (birds per km) in marine habitats in British Columbia. Data from appendix 

 1 (see text for selection criteria). Mean s.d. shown, and (in parentheses) range and sample size 



N. and W. Queen Charlotte Islands 



E. Queen Charlotte Islands 



2.7 2.6 

 (0-5.2: 3) 



5.4 4.4 

 (0.9-16.0; 11) 



1.1 1.2 



(0-4.0: 12) 



IM^^M^^^^^ 



2.6 0.2 

 (2.3-2.9; 6) 



Northern and Central Mainland 



Southern Mainland 



Gulf Islands and S. Strait of Georgia 



1.41.9 

 (0-4.2; 5 



2.1 1.8 

 L6-4.1;3) 



2.7 1.7 

 (0.4-4.9; 8) 



areas (12.9 birds/km 2 ), relative to more nearshore channel 

 and fjord habitats (6.3 and 0.2 birds/km 2 , respectively). Carter 

 and Sealy (1990) reported lower densities of murrelets in 

 mid-channel, than in shallower, sheltered nearshore waters 

 in Trevor Channel, Barkley Sound. High densities were 

 associated with sheltered, shallow nearshore water and at a 

 sill at the mouth of the channel, where a thermal front was 

 frequently found (Carter 1984). 



I summarized data from Carter's (1984) grid and transect 

 surveys in and near Barkley Sound to show the distribution 

 of Marbled Murrelets relative to distance from the shore (fig. 

 5). The distances to shore were estimated from the midpoints 

 of each of 12 sample blocks in Trevor Channel and from the 

 midpoints of each of 16 transect segments in Trevor Channel, 

 Imperial Eagle Channel, and the open sea off Cape Beale 

 (collectively referred to here as the Cape Beale transects). 

 The transect densities were converted from birds/km as given 

 by Carter (1984) to birds/km 2 by assuming a 250-m transect 

 width on each side of the boat (the same as used in the grid 

 surveys). Murrelets in this area were strongly aggregated 

 within 1 km of the shore, with highest densities 100-600 m 

 offshore {fig. 5). Similar results were found in grid surveys 

 in 1992 and 1993 (fig. 5 inset). Overall, the 1980 data fitted a 

 negative logarithmic curve: 



density [birds/km 2 ] = 2.438 d' 1 356 



(r 2 = 0.735, n = 28, P < 0.001) 



where d is distance from shore in kilometers. This curve is 

 plotted in figure 6 for values of d > 100 m. The density for 

 d < 100 m was assumed to be 7 birds/km 2 from the data in 

 figure 5. This model probably applies only to more 

 exposed shores of western Vancouver Island. 



A strikingly different distribution pattern emerges from 

 surveys made in Laskeek Bay, Queen Charlotte Islands, from 

 1989 through 1993 (see appendix I for references). Two 

 types of transects were sampled repeatedly: shoreline transects 

 within 400 m of the surfline and linear transects in open 

 water and among islands, up to 3 km from land. The open 

 water transects often had similar or higher densities than the 

 nearshore ones, and there was considerable variation within 

 and between seasons (fig. 7). The variability in these data 

 emphasizes the need for caution in interpreting distribution 

 and census data based on only one or a few seasons. 



In 1992 the positions of all Marbled Murrelets in open 

 water at Laskeek Bay were plotted on a chart to the nearest 

 200 m and, at this fine scale, show the widespread and variable 

 distribution of murrelets relative to the nearest land (fig. 6). 

 The pattern in these sheltered waters is quite different from 

 that of the more exposed outer Barkley Sound area shown by 

 the logarithmic model (fig. 6). These data clearly illustrate the 

 problems in estimating total densities for any marine area 

 from extrapolations of one or more shoreline transects. The 

 distribution pattern for open and exposed water is quite 



USDA Forest Service Gen. Tech. Rep. PSW-152. 1995. 



301 



