Asymmetrical force production in the maneuvering flight of pigeons

DOUGLAS R. WARRICK,1,3 KENNETH P. DIAL,1 AND ANDREW A. BIEWENER2

ABSTRACT.-Downstroke force produced by Rock Doves (Columba livia) as they negotiated an obstacle course was measured using in vivo recordings of delto-pectoral crest strain. During this slow (

THE GREAT DIVERSITY OF SPECIES within all taxa capable of flight suggests that the behavioral plasticity afforded by flight allows the exploitation of a wide variety of habitats (Norberg 1990). The directness and speed of flight allow great distances to be covered efficiently and quickly such that flying animals can respond to changes in food supply, climatic conditions, or, in more immediate situations, predation pressure. However, speed isn't everything; studies of flight performance (as inferred by external morphology) suggest that the spatial characteristics (e.g. clutter) of a species' habitat exert selective pressure on maneuverability (Norberg 1981, Norberg and Rayner 1987). The inference of maneuvering performance by external flight morphology (e.g. wing loading, aspect ratio) requires a steady-state assumption (i.e. continuous lift production and nonflapping wings). On the contrary, given the enormous variety of bird species that rarely glide, particularly during maneuvering (e.g. small passerines), the vast majority of maneuvering flight must take place during flapping flight. Therefore, our understanding of the relationship between flight morphology and the ecology of the animal must be limited. Furthermore, the proximal physical mechanisms for changing direction in slow, unsteady flight are not understood, which prevents clearly relating this type of maneuvering performance to morphology beyond the requirement of high available mass-specific power.

The mechanisms for changing direction in a steady-state turn are fairly simple: disparate forces produced by the wings cause the bird to roll into a bank (henceforth, an initiating force asymmetry), redirecting lift toward the desired direction of flight. In this type of turn, after a reversal of the initial force asymmetry to halt the rolling momentum of the bird (henceforth, an arresting asymmetry), no further force asymmetry is needed to maintain the bank once it has been established, and the bird turns at a constant rate. The process of changing direction during nonsteady-state (flapping), slow flight is less clear, although presumably it is of critical importance; most birds must be able to maneuver precisely during takeoff and landing. Using two-dimensional light film, Dial and Gatesy (1993) found that pigeons flying at low speed through an obstacle course appear to create a bank using angle-of-attack asymmetries and then fly symmetrically through a turn for several wingbeats. If true, the maneuvering strategy of pigeons in slow flight is the same as that employed in a steady-state turn: they create a bank and allow the redirected lift of downstroke to pull them steadily through the turn. However, kinematic asymmetries may be difficult to discern on two-dimensional film, and small wing asymmetries and changes in yaw and bank may have gone undetected during each wingbeat. Such a slow-flight turn would be produced by a series of small changes in direction (henceforth, a saltatory turn) rather than by establishing a bank and subsequently flying symmetrically through the turn (i.e. a symmetrical turn). The purpose of the present study was to use paired in vivo force recordings of the pectoralis, based on strain recordings of the delto-pectoral crest, to examine the patterns of downstroke force production in pigeons flying through an obstacle course, and thereby determine which turning strategy (symmetrical or saltatory) they employed during slow, maneuvering flight.

MATERIALS AND METHODS

Bird training, flight corridor, and cinematography.Three Rock Doves (Columba livia; hereafter "pigeons") captured in Missoula, Montana, were trained to fly from the hand through a hallway corridor (2.5 x 3 x 20 m) to a perch. Four clear (to allow the flights to be filmed) acetate barriers (1 m wide x 2.4 m long, 0.08 mm thick) were suspended at 4-m intervals within the corridor such that the pigeons needed to make four alternating turns to negotiate the flight course (Fig. 1). The takeoff and landing points were placed far enough from the first and last barriers, respectively, to allow the pigeons to initiate/complete the first and last turns without being influenced by the need to accelerate/decelerate to land. The obstacle course was arranged such that either type of turn (saltatory or symmetrical) was physically possible, with the maximum radius of 12.6 m for a symmetrical, constant-rate turn around any one of the barriers. Assuming a modest lift coefficient of 1.2 (Norberg 1990), a turn of such radius could be executed by a gliding pigeon in a 45deg bank. During training, the acetate barriers were marked with strips of high-visibility tape to allow the birds to see and learn to avoid the barriers; as the birds became familiar with the course, the tape was gradually removed. The birds were considered ready for experimental trials when they could fly through the entire course without striking the barriers. Four photocells were mounted on the ceiling at the medial edge of each barrier to mark a bird's position as it passed through the assumed middle of each turn. A 16-mm high-speed film camera (RedLake Laboratory Lo-Cam; Kodak 7250 Ektachrome) operating at 150 frames s^sup -1^ recorded a posterior view of each trial, allowing two-dimensional kinematic reference and analysis.