FISH AND WILDLIFE TECHNICAL REPORT 25 



ing to identify each bird to species. Usually enough birds 

 are identified to species to provide a guide to interpreting 

 unidentified birds. If birds are being missed because the 

 observer is concentrating on identification, the observa- 

 tions should be coded as "general observations" and not 

 used to develop indexes of abundance. Studies devoted 

 to single species can make use of our techniques, but the 

 surveys cannot be pooled with data relating to total 

 seabirds. 



Concentrate on the actual count zone and do not spend 

 much time scanning outside of that area. Uncommon 

 sightings and flocks observed outside of the count zone 

 should be recorded, but they should not be actively sought, 

 as this can result in birds being missed within the count 

 zone itself. Perception of the transect' s width narrows with 

 distance, and it does not appear to be as wide at 1 ,000 m 

 as it does at 100 m; take this into account when deciding 

 which birds should or should not be recorded as within 

 the transect zone. One often has to wait until a bird is 

 directly abeam of the ship to decide if it is within the 

 transect zone but be cautious, for by that time the bird 

 may have moved out of the zone in trying to avoid the 

 ship. Record all sightings of marine mammals and of bird 

 flocks greater than 1,000, whether they are in or out of 

 the count zone. In the case of large flocks over large areas 

 try to make one estimate of total flock size even if it may 

 extend for several miles. 



Theoretically, one is attempting to obtain an instan- 

 taneous count of birds within the count zone rectangle 

 (Fig. 1). Birds that enter the count zone from behind the 

 ship (area already surveyed) are not counted, while those 

 that enter or leave in front of the ship are counted. There 

 is one exception to the rule about not counting birds enter- 

 ing the count zone from behind the ship. If traveling birds 

 are moving in the same direction as the ship, then those 

 in the count zone during instantaneous counts should be 

 recorded. 



Large numbers of traveling birds present a special prob- 

 lem. If the observer counted all the individuals flying 

 through the count zone, density indexes would not only 

 be greatly exaggerated, but would reflect birds using the 

 air corridor over the water rather than being associated 

 with the water itself. To reduce this particular bias, we 

 have a special method of counting traveling birds. 



When there are traveling birds passing through the area 

 each individual is not recorded. Instead, we make instan- 

 taneous counts of these birds within successive sections 

 of the count zone (Fig. 1). The number and size of in- 

 stantaneous count zones depends on the maximum distance 

 at which all of the traveling birds can be detected and the 

 speed of the ship. The area covered by all of the instan- 

 taneous counts added together always equals the total 



transect count zone. For example, during a 10-min tran- 

 sect at a speed of 10 kn, the ship would cover a total 

 distance of 3,087 m (Table 2). For large flying birds, we 

 would take three instantaneous counts each covering an 

 area extending about 1 ,000 m ahead of the ship and 300 m 

 to one side. One count would be taken at the start of the 

 transect, one at 200 s (ca. 3.3 min) into the transect, and 

 one at 400 s (ca. 6.7 min) into the transect. The three 

 counts added together would be our estimate of the 

 number of traveling birds in the entire transect at any one 

 time. Value judgements as to distance and whether to 

 include this or that bird become easier and more trust- 

 worthy with experience. For smaller flying birds, such 

 as storm-petrels (Hydrobatidae), an instantaneous count 

 zone of 300-500 m is usually more appropriate. Instan- 

 taneous counts to 300 m ahead of a ship moving at 10 kn 

 would be taken at approximately 58-s intervals. If birds 

 are sitting on the water or there are other indicators of 

 position, instantaneous counts can be judged by these ob- 

 jects rather than by the time and speed of the ship. In sum- 

 mary, instantaneous counts are an attempt to obtain a 

 single picture of traveling birds within the total count zone 

 at any one time by putting together a series of smaller 

 pictures. 



Occasionally a judgement will be required as to whether 

 to use instantaneous counts for a large flock of foraging 

 birds. Normally such flocks are counted only once when 

 first observed, but if the flock is larger than the total 

 transect then it may be more appropriate to treat them as 

 traveling birds and use the instantaneous count method. 

 The distance from the observer to the end of the transect 

 at various ship speeds can be obtained from Table 2. 



A situation requiring special treatment is that of a large 

 flock of birds being deflected in front of the ship; for ex- 

 ample, 10,000 short-tailed shearwaters, all in one flock, 

 streaming along the side of the ship and then across the 

 bow. The flock is continuously passing in front of the ship 

 because it is being deflected forward (Fig. 1). Such a flock 

 should only be counted once (i.e. , in the first instant count) 

 and then ignored in all future counts. 



Land birds and flocks of shorebirds that are obviously 

 just passing over the area on migration or moving between 

 two distant points are handled differently than marine 

 birds. By using proper coding techniques (see Appen- 

 dix B), these sightings can be included in the data base 

 without influencing density indexes. 



Estimating numbers is a major source of bias in surveys. 

 Before going into the field, practice estimating large 

 numbers of objects such as beans on a table or birds in 

 pictures (Arbib 1972). Most field observers estimate the 

 number of birds in large flocks by counting in 10's or 

 100' s. This requires the observer to maintain a firm mental 



