Formerly www.swift.utigges.il.eu.org

Counting Common Swift (Apus apus) Populations

New discoveries in the migratory, as well as spatial and temporal, behaviour of the Common Swift in large cities, at last enables precise counting of their populations. It is now known that there are set periods, or "time windows", when one can count separately the potential breeders and the non-breeders. Armed with the exact arrival times of the Swifts in their breeding areas, one can now set precisely the times for a successful census.

The following analysis of Swift behavior permits census-taking using a time window method.

Swifts arrive in their breeding areas in four waves. The mature birds arrive in the first three waves at the start of the breeding season. The immature yearlings arrive in the fourth wave, about mid-way through the breeding season.

In the mornings and evenings all Swifts fly within a set colony territory, usually extending from about 90,000 m² to 120,000 m² . Within square territories the borders are approximately 300 m x 300 m; within triangular territories they are approximately 450 m – 450 m – 450 m.

Breeding Common Swifts

Potential breeding Swifts may be counted from seven to fourteen days after the arrival of the second or third wave. At this time all the birds are within the breeding areas, and in general they have not yet become intensely occupied with courtship and egg-laying.

The time window for counting depends on the time of twilight, and in turn is related to the geographical location. In Jerusalem (31º 52' N, 35º 13'E) the time window is between 0700 and 0800 hours Israel Standard Time (IST) or else between 1730 and 1800 hours IST.

In Berlin (52º 28' N, 13º 24' E) the time window is 0900 to 1000 hours Central European Summer Time (CEST) or between 1900 and 2000 hours CEST.

At these times the Swifts are out of their breeding holes, and circling the colony territory in flight.

At least two counts should be made, preferably more, so that birds missed the first time (either because they are in their nest holes or out of sight) can be picked up on the second count.

Non-breeding Common Swifts

The census of non-breeding Swifts should be taken in the last third of their stay. The best time for this is in the 5^th and 6^th weeks before departure. In Berlin this is between July 1^st to 15^th. The census should be timed for the periods when the birds are either returning to the colony, after being out all night, or else prior to assembling at the colony for the night out. In Berlin the appropriate time windows would be 0800 to 0900 hours, and 2000 to 2100 hours CEST.

At these times the breeding Swifts are occupied exclusively with the search for food for their young, and do not participate in the territorial and social flights of the non-breeding colony members. The only breeding Swifts likely to be encountered during this census would be those bringing food to the nest, and leaving to fetch more.

Taking a census in a city

Swift populations within well-defined areas may be counted, if a reasonable number of helpers can be assembled. Since Swifts are exclusively aerial, (apart from when they are nesting, and sometimes roosting), and perform territorially at certain times of the day, they may be counted by making a succession of observations at pre-set points within the area to be covered, using a progressive wave technique.

The helpers should be positioned at exact points set on a vertical axis, about 200 m to 300 m apart from each other, depending on the availability of suitable positions (e.g. road crossings). The census should take place at fixed times, and all Swifts in flight in the marked area must be counted.

As it is quite feasible that either the whole colony, or parts of it, are flying in a different part of the territory, the helpers must make successive counts from the same positions 5 and 10 minutes after the first.

Then the helpers move forward to the next counting points, set laterally about 200 m to 300 m from the first set, and the counting recommences. Using this technique quite large areas can be covered reasonably quickly.

(Source: Tel Aviv Map and Guide 1997, changed)

In the mornings and evenings Common Swifts establish their colony territories by circling above them continuously. To count them one establishes a vertical chain of set points about every 200 m to 300 m. At the same fixed time the Swifts are counted at each set point. In addition one makes a note of their direction of flight. Successive counts are necessary, 5 and 10 minutes after the first. Then the counters proceed laterally to the next set of counting points, and the counting procedure starts again. The map above shows an example census area and the positions needed by a counting team of four.

Taking a census in larger cities

Within bigger cities the best way of counting Swifts may be to use the spot check method. To do this one divides the area by building type / use (e.g. areas of old buildings, areas of new buildings, apartment blocks, terraces, semi-detached and detached houses).

By making a random selection one chooses areas for the census. The results from these randomly selected areas are used to generate data for the whole city. This method was used successfully in the Berlin Census for 2002.

Taking a census in smaller towns

Initial studies suggest that the Common Swift population in smaller towns and villages behaves in the same way, both spatially and temporally, as in the larger built-up areas. However, as precise research has yet to be conducted, more work is needed before counting in such areas can be performed in confidence using the methods suggested in this article.

Analysis

In evaluating the results of the counts the following need to be taken into consideration:

• The choice of time windows follows the median dates for specific Common Swift activity. However counters should be aware that variations in Swifts behavior may occur.

• Efforts must be made not to double-count individual Swifts.

• Care should be taken to conduct counting only in fine weather. Do not count during, or just after, a period of cold or wet weather, as the Swifts may have been driven away from their usual areas by the bad weather.

• The percentage of breeders likely to be accidentally included in the census of non-breeding Swifts is very small. In the Berlin census of 2002 the number was less than 1 %.

• As we lack valid data on the relative abundance of breeders / mature non-breeders / immature yearlings the results of the census should be divided into "potential breeders" and "non-breeders" only. A survey of ringed Swifts conducted by Erich Kaiser at a Common Swift colony near Frankfurt am Main produced the following results:

Breeding birds 45 %

Mature non-breeders 20 %

Immature yearlings 35 %,

but not all the birds in the colony could be examined.

© Ulrich Tigges 2003

References: APUSlist: 0060, 0061, 0067, 2741, 2856, 2867