Strong and others 



Chapter 32 



Distribution and Population Estimates in Oregon 



25 

 20 

 15 

 10 



ll 



I I 



I 



I. 



I 



5101 1230000102000001210050020 12 811909364321010101 10212020021 



* 15 



o 

 m 



<T 



!T 10 



.1. 



2010112110201000203410010001 1513129 011131717235 00110002123401003 



Figure 4 Average numbers of Marbled Murretets at sea counted from shore in Oregon. Numbers on x axis represent the number of 

 counts within a 1 0-km section of coast. Arrows indicate division between regions. Refer to fig. 3 for locations along the Oregon coast. 



the largest in 1992, groupings of murrelets was also very 

 similar in 1992 and 1993 (fig. 8). 



Distance from the boat at which murrelets were reported 

 was similar among years, except in the 20 to 50 m range (fig. 

 1). It is likely that this resulted from bias in reporting distances 

 in 1992 when we had predetermined our strip width to be 50 

 m for density estimates. In 1993 there was no such 

 presupposition and we took care to visually calibrate our 

 estimates with a 50 m measured buoy line and among ourselves. 

 Based on the curve in Figure 1 and on density computations 

 for various strip widths (fig. 2) we selected a strip width of 50 

 m on either side of the boat (100 m). This strip width included 

 74 percent of all birds seen in 1992 and 64.2 percent in 1993, 

 not including flying birds. Fewer birds were reported closer 

 than 20 m since they usually took evasive action at greater 

 distances. Marbled Murrelets dove in avoidance of the boat 

 at a mean distance of 26.5 m (s.d. = 1 8.6 m), and they flew in 

 avoidance at 42.6 m (s.d. - 36.1 m). 



Population Estimates 



Comparison of Aerial, Vessel, and Shore-Based Estimates 



Vessel estimates using line or strip transect analyses 

 produced far higher estimates than air or shore-based surveys 

 (table 1). All methods used densities calculated for 1 km 2 

 in the estimates (table 3) except for the central region, 

 where there was information on offshore distribution (fig. 

 5). For the central region, 1 km 2 densities were halved to 

 estimate only the number out to 500 m, and estimates from 

 offshore sample densities, in 500-m blocks, were summed 

 and added to the estimate (the offshore component is shown 

 separately in Table 1). We added the same proportionate 

 number of birds to air and shore-based estimates in the 

 central region as were added to vessel estimates in accounting 

 for offshore distribution. Differences between estimates, 

 then, were due to differences in mean densities of birds 

 detected with each method and year (table 3). Of the three 

 survey methods, vessel transect data had the highest 



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



345 



