Strong and others 



Chapter 32 



Distribution and Population Estimates in Oregon 



Offshore Transects 



Transects sampling offshore waters were grouped in 

 500-m increments of distance from shore, and each sampling 

 day was treated as a replicate within the groups. The data 

 within each 500-m group were then modeled with a detection 

 curve for line transect and a density calculated using the 

 DISTANCE program, or summed and divided by transect 

 length for strip transects as described above. In this way, a 

 separate density estimate was calculated for each 500-m 

 increment offshore for both line and strip transect methods 

 in 500-m by 1000-m blocks. These densities were then 

 multiplied by the length of the central region for independent 

 abundance estimates within each 500-m increment offshore. 

 The sum of these abundance estimates were added to the 

 central region when incorporating birds offshore in overall 

 population estimates (this offshore component is shown 

 separately in table 1). 



Aerial Estimates 



Similar strip transect methods as were used on the vessel 

 were used in aerial surveys. Each observer's results were 

 treated as a separate transect (sometimes only one observer 

 could conduct transects due to glare on one side, for example), 

 so total strip width was 50 m and the number of transect 

 samples was greater. Densities were calculated by dividing 

 the total number of murrelets seen by each observer in a 

 region on a transect by the length of that region. Densities 

 were multiplied to measure square kilometers, and then 

 multiplied by the length of each region for population estimates 



as with vessel surveys. For the central region the proportion 

 of birds occurring over 500 m from shore, based on vessel 

 offshore transect data, were added to the region's estimate. 



An independent estimate was calculated for each day, 

 and these data were then averaged for the regional estimate, 

 as with vessel estimates. 



Shore-Based Estimates 



To summarize shore-based observations, we assumed a 

 145 angle of view (given a 150 m wide surf zone and 

 setback from the shoreline) and measured an approximate 

 viewing limit of 1.2 km out to sea, which gave a scanning 

 area of roughly 2 km 2 . To compensate for low viewing angle 

 over surf, we halved the scanning area to 1 km 2 as an actual 

 survey area when computing densities. The average number 

 of murrelets counted from all points in each day was multiplied 

 by the length of the regions coastline for an independent 

 daily estimate, as was done for air and vessel transects. 

 These values were then averaged for a regional population 

 estimate. The proportion of birds greater than 1 km offshore 

 from the vessel offshore strip transect data were added to 

 central region estimates as with aerial estimates. 



Field Effort 



Field work was carried out from 1 June to 15 August in 

 1992 and from 10 May to 1 August in 1993. Our research 

 effort was primarily devoted to vessel surveys, and most of 

 the vessel transects took place in the central region, between 

 the Siletz and Siuslaw rivers (table 2). 



Table 2 Summary of survey effort for Marbled Murrelets off the Oregon coast in 1992 and 1993. Initial training 

 transects and transects fragmented by weather or data recording errors were discarded prior to analyses. \'rssel 

 surveys were separated into extensive coastline and offshore distribution transects 



1 Air survey strip width was only 50 m wide as each observer's data was considered an independent survey (flights 

 actually covered half the listed km). 



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



343 



