2. Peaks in the morning and late afternoon. 

 Juan Fernandez and Black-winged Petrels 

 showed this activity cycle most clearly. 



3. Peak in the middle of the day. Leach's 

 Storm Petrels, Red-tailed Tropicbirds, 

 Brown Noddies, and White Terns appeared 

 most often in the middle of the day. 



4. Peak activity at noon or late morning and 

 again late in the afternoon. This pattern 

 was followed by Sooty Terns, Bulwer's 

 Petrels, and, to a lesser extent. Wedge- 

 tailed Shearwaters. 



It is difficult to assess the meaning of these 

 data in terms of what birds are actually doing. 

 Theoretically, if movement is sporadic and un- 

 motivated, fewer birds will be observed and 

 more will be sitting on the water; conversely, 

 if there is a heightened urge to feed or travel, 

 activity should increase and a greater number 

 of birds should be seen. The similarity in the 

 activity cycles of species related systematical- 

 ly or by similar feeding habits, e.g., Juan 

 Fernandez and Black-winged Petrels, lends 

 some support to the data as a measure of avian 

 activity. 



Environmental Influences 



A preliminary analysis of how wind direction, 

 wind speed, air temperature, surface water 

 temperature, and surface salinity affect the 

 density of the more common sea bird species 

 is presented below. The environmental data 

 are arbitrarily applied to all sightings made in 

 the period from one environmental observation 

 to the next. Because environmental observa- 

 tions were made every 3 hours or less, the 

 greatest distance a bird might have been seen 

 from the site of the environmental observation 

 within which it was included was about 30 miles. 

 Environmental observations were grouped ar- 

 bitrarily to yield from 5 to 10 incremental 

 categories. Only observations taken between 

 hours 0500 and 1900 were tallied. The number 

 of birds recorded under each category was 

 divided by the number of environmental obser- 

 vations made in that category. Since the en- 

 vironment was sampled every 30 miles the 

 resulting figure of birds per sample is equiva- 

 lent to birds per 30 miles of observation. 



Wind direction . — Relative densities of birds 

 at various wind directions (table 7) showed no 



conclusive relations, although the relatively 

 high numbers of tropicbirds and Great Fri- 

 gatebirds observed when the wind was from the 

 west or southwest may be significant. A west 

 wind could blow these species to the study area 

 from the main or leeward Hawaiian islands. 



Wind speed . --The numbers of birds observed 

 at various wind speeds is believed to be a mea- 

 sure of the effect of wind on the observability 

 of the birds, rather than an indication of move- 

 ment of birds to avoid or to take advantage of 

 the winds. Figure 32 shows densities of the 

 more common species at various wind speeds 

 in increments of 4 knots. Several basic pat- 

 terns are evident. The first, typified by Leach's 

 Storm Petrel and Bulwer's Petrel, has two 

 peaks--a high peak when the wind is calm, or 

 nearly so, making small, low-flying birds vis- 

 ible at a greater distance, and a lower peak at 

 the greater wind speeds, when they arc higher 

 than normal (reach the high point in an undula- 

 ting flight). The second pattern, typified by 

 Sooty and Slender-billed Shearwaters, has a 

 single peak at medium to high winds. As winds 

 increase, these species arc higher and are vis- 

 ible at a greater distance. Albatrosses are 

 seen most often during strong winds because 

 they arc higher with increasing wind speeds and 

 their large size makes them visible at a dis- 

 tance. Tropicbirds are observed most often 

 during light winds; they are most often seen 

 sitting on the water or are spotted directly 

 above the ship because their call attracts the 

 observer's attention. Waves obscure sitting 

 birds from view, and the noise of the wind 

 makes it difficult to hear them calling. 



Air temperature . --Avian densities calculated 

 for each of seven categories of air temperature 

 at 2° increments between 16° and 30° C. are 

 presented in figure 33. Several species show 

 remarkably well-defined "preferences" for 

 certain air temperatures. Even though the 

 numbers of environmental samples at the low- 

 est and highest categories were low (three and 

 nine, respectively), it is probably significant 

 that only albatrosses were seen at temperatures 

 in the lowest category and that they were most 

 abundant at those temperatures. No albatrosses 

 were observed at the temperatures in the high- 

 est category. 



Several species were observed in greatest 

 densities at 26° to 27° C. These included the 

 two most abundant breeders in the mainHa- 



8<t 



