As mentioned, swallows and swifts are not suited for the traditional survey methods which estimate density, such as spot mapping, line transects, or fixed radius point counts. Most studies, even if they collect data on one of these species, ignore such data in the analysis because of extreme data variation. For instance, Manuwal (1991) used the variable radius point count method to determine which bird species showed a dependency on old growth forests in the Pacific Northwest. He could not use the data collected for Vaux's Swifts because "of its wide ranging behaviour" (p.165). There are several reasons why traditional techniques are inefficient for swallows and swifts.
First, there is a problem of a mismatch in scale. Traditional surveys are usually conducted on areas no larger than 20 ha, while the foraging area for swallows can be much greater. For example, the Tree Swallow has been found foraging up to 100 km from its nest site (Robertson et al. 1992). These mismatches result in samples taken within only a fraction of the foraging area, so the probability of one of the birds being in that area at any one time is quite low. Second, for most of these species, singing for solicitation or defence of mates occurs before dawn (Bent 1942). Since most surveys begin at or just shortly after dawn, this key activity is missed. Third, swallows and swifts begin foraging about four hours after sunrise, once their prey have become airborne. Robertson et al. (1992) report that Tree Swallows often do not leave the nest site until midday during incubation. Again, the timing of the traditional surveys during the morning chorus is not suited for this group. Fourth, in a general sense this group can be considered colonial. Some species show this behaviour more than others, but it is commonly understood that they have a clumped foraging distribution. The traditional surveys are designed for species which are regularly distributed in space. Finally, surveys are rarely conducted in urban environments, yet these are the areas where the many of this inventory group forage and breed (Weller and Francis 1987).
Because swallows and swifts are not suited for traditional surveys, other methods must be designed. One of the keys to a successful survey design is an understanding of the birds' behaviour. The survey designs recommended in this manual take advantage of two characteristic behaviours: group foraging and coloniality.
Group Foraging
Unlike breeding pairs of other small passerines (e.g., warblers and sparrows) that feed in exclusive territories, where food is more or less uniformly distributed, swallows and swifts tend to congregate where food is abundant. Due to the ephemeral nature of their prey base, and because the precise location of good feeding sites varies from hour to hour, swallows and swifts use each other as clues to finding food (Gill 1990). Consequently, large mixed species flocks can be observed foraging over open areas where the food supply is abundant.
Coloniality
Whereas stable food resources and defensible areas promotes territoriality, unstable food resources and indefensible areas promote coloniality (Gill 1990). Consequently, the transitory nature of the prey base which promotes flocking, also promotes coloniality. For example, Emlen and Demong (1975) found that Barn Swallows keyed on the location of aerial insects by following the flight line of other swallows which returned to the colony with food.
The nature of the food base is not the only reason for coloniality and flocking. Another important factor is reduction in risk of predation. Swallows which nest in large colonies or forage in large flocks, can reduce their risk of predation in any of three ways: 1) confusion of the predator because there are too many birds for it to focus on; 2) reducing the probability of predation per individual in larger flocks; and/or 3) increasing the chance of a predator being detected. Consequently, the behaviours associated with flocking and coloniality determine the type of survey methods which can be deployed.
Table 1 outlines recommended methods for inventory of swallows and swifts in British Columbia at three levels of intensity. The survey types listed are unlimited radius point count surveys (presence/not detected and relative abundance), and active nest counts (relative abundance). Mark-resight survey protocol for swallows and swifts (absolute abundance) is experimental at this time, and is not recommended for use as a standard inventory method. The two methods recommended for relative abundance measure different attributes of the population. Active nest counts are appropriate for barn, bank, cliff, violet-green and possibly tree swallows. However, unlimited radius point counts are recommended for swifts because of the hidden nature of their nests. Potentially all swallows and swifts could be enumerated during an unlimited radius point count survey (dependent on habitat and geography).
Table 1. Recommended inventory methods for swallows and swifts in British Columbia at 3 levels of intensity.
| Level of Intensity | Recommended Method(s) |
| Presence/Not Detected | |
| Relative Abundance | |
| Absolute Abundance |