| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Behavioral Ecology Vol. 12 No. 5: 607-611
© 2001 International Society for Behavioral Ecology
Sexual differences in tailwind drift compensation in Phoebis sennae butterflies (Lepidoptera: Pieridae) migrating over seas
Department of Zoology, South Parks Road, University of Oxford, Oxford OX1 3PS, UK, and Smithsonian Tropical Research Institute, Apdo. 2072, Balboa, Republic of Panama
Address correspondence to R.B. Srygley at the University of Oxford. E-mail: bob.srygley{at}zoo.ox.ac.uk .
Received 15 November 1999; revised 4 September 2000; accepted 22 November 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
One prediction derived from optimal migration theory is that migrating animals that maximize their flight distance on a given amount of energy will decrease their airspeed in a tailwind and increase it in a headwind. To test this in a migrating butterfly, I followed male and female cloudless sulfur butterflies Phoebis sennae (Pieridae) migrating from Colombia toward Panama over the Caribbean Sea. P. sennae headed westerly over the Caribbean Sea in the morning and then turned southeasterly to head downwind in the afternoon. Changes in heading and track directions of P. sennae were not related to changes in the position of the solar azimuth. As predicted from optimal migration theory, flight velocities of females decreased in a tailwind to minimize energy consumption. However, males did not show any compensation for tailwinds. Females are minimizing energy consumption, whereas males may be minimizing the time to reach the destination site in order to maximize matings with newly arrived or newly emerged females. Orientation of females changed before that of males, presumably because their greater reproductive load imposed greater flight costs and limited flight fuels.
Key words: Caribbean, cloudless sulfur butterfly, drift compensation, flight, migration, orientation, Phoebis sennae, tropical butterflies, sexual dimorphism.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Theoretical analyses of the energetic costs of flight have identified optimal strategies for aerial bats, birds, and insects that migrate long distances (reviewed by Alerstam and Hedenström, 1998
;
Richardson, 1991). "Long
distance" is obviously relative to the size of the organism, and here I
apply it to distances over which a large proportion of the energy reserves of
the organism are consumed.
There are four clearly defined velocities that a migrant may adopt to
optimize different aspects of its time and energy budgets. The minimum power
velocity, Vmp maintains the animal aloft for the longest
period, and the maximum range velocity, Vmr maximizes
distance for a minimum required energy. Maximum velocity,
Vmax, which is limited by the power available for flight,
minimizes flight time to the destination site when there is no feeding en
route. When time for energy deposition is budgeted, the minimum time velocity,
Vmt, is less than Vmax and greater
than Vmr (for recent reviews, see
Alerstam and Hedenstrom, 1998;
Srygley and Oliveira, 2001).
The maximum range velocity is derived graphically by drawing a tangent from
the origin to the U-shaped power curve for flight. However, the origin of this
tangent will shift negatively with the velocity of the tailwind (and
positively with a headwind; Pennycuick,
1978) such that the maximum range power is greater in a headwind
and lesser in a tailwind relative to its magnitude in a still wind
(Figure 1). A qualitatively
similar change in airspeed with a change in tailwind is predicted for the
minimum time velocity (Alerstam and
Lindström, 1990). For migrating
birds, adjustment of airspeed for tailwinds has been demonstrated repeatedly
(see citations in Alerstam and
Hedenström, 1998). Honeybees
(Apis mellifera) reduced airspeed in a tailwind and increased
airspeed in a headwind when flying 250 m across a lake
(Heran and Lindauer, 1963).
|
Adjusting flight velocity to maximize the range that an insect is able to
fly is likely to be under strong natural selection in migratory butterflies
flying over the sea without access to nutrients. I investigated changes in
airspeed relative to varying tailwinds in males and females of the cloudless
sulfur butterfly Phoebis sennae migrating over the Caribbean Sea.
Because I lack data necessary to generate a power curve for P.
sennae, I based its shape on that derived for migrating moths Urania
fulgens (Dudley and DeVries,
1990) and assumed that it is J-shaped at velocities greater than
the minimum power velocity.
The use of a time-compensated sun compass or magnetic compass is also
likely to be important when flying over the sea
(Oliveira et al., 1998). To
determine whether the orientation of the butterflies depended on the sun's
position, track directions of migrating P. sennae were associated
with the direction of the solar azimuth over the course of the day.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study organisms
Phoebis sennae Linn. (Lepidoptera: Pieridae: Coliadinae) is a regular migrant in the Caribbean (Brown and Heineman, 1972
), Central America (Srygley RB, Oliveira EG, and
Dudley R, unpublished observations) and the southeastern United States
(Walker, 1991;
Walker and Littell, 1994).
During late November to mid-December 1997, P. sennae were observed
migrating over the Caribbean Sea near Colombia from Cartagena to
Tolú (90 km south of Cartagena;
Figure 2) and as far north as
Santa Marta (170 km northeast of Cartagena; Zea S and Franke R, personal
communication). Butterflies were also observed flying over the Caribbean Sea
62 km southwest of Cartagena (Londono C, personal communication).
|
Males and females are sexually dimorphic in body size and coloration. Males
are lemon-yellow and generally smaller (n = 7, mass = 154 ± 24
mg; n = 6, winglength = 35 ± 1 mm) than the off-white females
(n = 3 females, mass = 173 ± 57 mg; n = 2, winglength
= 37 ± 2 mm; body size data from butterflies in lowland rainforest of
Panama). In Panama, larvae feed on Senna frondosa (formerly in the
genus Cassia) and Inga goldmanii (Leguminosae; Srygley RB,
unpublished data). Nitrogenous resources for spermatophore and egg production
are gathered in the larval stages, and lipids are acquired during larval and
adult stages (May, 1992). Both
sexes gather nectar at flowers, and the adult males puddle (Srygley RB,
personal observations), presumably to gather salts which are transferred to
females during copulation (Pivnick and
McNeil, 1987; Smedley and
Eisner, 1996).
Individual butterfly airspeed, track direction, heading, and local
wind
Phoebis sennae butterflies were intercepted while flying over the
Caribbean Sea and followed in a 25-foot fiberglass speed-boat powered by a
140-hp inboard/ outboard motor until an even pace was maintained parallel to
the flight direction. The pilot and I sampled butterflies flying over the
water between San Bernardo Islands and Baru Island approximately 15-20 km
offshore of South America on 7 and 11 December 1997, between San Bernardo
Islands and Tolú approximately 10 km offshore
on 8 December 1997, and approximately 15 and 10 km west (offshore) of San
Bernardo Islands on 9 and 10 December 1997, respectively. Sampled butterflies
remained within 0.5-8.5 m of the water surface and progressed forward steadily
without evasive behavior.
Boat heading was measured with a flux-gate compass (Raytheon heading sensor M 92649) mounted on the deck, approximately 1 m above the water line. Boat speed was measured with a transducer (Airmar P55/#20-039) on a transom-mounted paddle wheel. Apparent wind direction and apparent wind speed were measured with a wind vane and anemometer (KVH Quadro network speed/wind director) mounted together on a 0.5-m aluminum pole extending over the bow from a 2-m mast that was erected on the boat deck approximately 3-3.5 m above sea level. Data were integrated with a KVH Quadro NMEA (National Marine Electronics Association) concentrator, and true wind speed and true wind direction were calculated with a KVH Brain (4321). Positional coordinates, as well as speed and true course over ground, were collected from a deck-mounted global positioning satellite (GPS) receiver. Every 5 seconds, all data were stored in a palmtop computer (Hewlett-Packard HP200LX) with the date and time using a customized DOS BASIC program (Trimble A, unpublished application).
The navigation equipment was calibrated using standard techniques on the first morning when wind speed was 0-0.5 m/s. Across the speeds relevant to this study, mean boat speed was within 0.5 m/s of mean speed over ground (Figure 3). True wind speed was within 0.5 m/s of the wind speed measured with a Kurz hand-held anemometer when the boat was still, and error in wind direction was negligible.
|
While tracking a butterfly, I grouped potential landmarks into three categories: none, ahead (i.e., those in the track direction), and visible (i.e., others that were not directly ahead). The insect's height over the water was estimated to the nearest 0.5 m.
I calculated the butterfly's heading and airspeed using a wind-drift vector
analysis (after Srygley et al.,
1996) with minor modifications for measured ground speed rather
than measured airspeed. Tailwind velocity is the wind vector component along
the track direction (negative for a headwind). For each individual, data were
averaged for the sampled interval (butterflies were followed for 40-430 s).
The female followed for the least amount of time (40 s) was lost in the glare
off the sea, whereas the male followed for the least amount of time (50 s)
encountered another butterfly and both circled upward until lost from
view.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Over a similar range of tailwind velocities (Figure 4), the slopes for the regression of airspeed on tailwind were significantly different among the sexes (ANCOVA, p = 0.011). Airspeed declined significantly with tailwind velocity for female P. sennae (n = 19, p =.002), whereas it was not related to tailwind velocity for males (n = 20, p = 0.68). For the males, the 95% confidence limits (CL) for the slope were -0.13 and 0.09, of which the lower value does not eliminate a small degree of tailwind drift compensation.
|
Males and females were sampled at similar frequencies between morning and
afternoon (
2 = 0.05, df = 1, p =.82) and at similar
frequencies with relation to the visibility and orientation of landmarks
(2 = 2.67, df = 2, p =.26). On two dates (8 and 9
December), only males were sampled, including the two males measured in
tailwinds that exceeded 4 m/s (Figure
4). The results were robust to the exclusion of males sampled on
8-9 December. The intercept for airspeed was not different from that when none
of the males were excluded (4.24 m/s relative to 4.24 for all males), and the
regression of airspeed on tailwind velocity was not significantly different
from zero (n = 15, p =.52). Although the 95% CLs for the
slope were broader due to the smaller sample size, the lower CL was less
toward the predicted decline in airspeed than when no males were excluded
(-0.10 relative to -0.13 for all males). Hence, the difference between males
and females in their responses to tailwind drift was not due to a difference
in spatial or temporal sampling.
The sexes did not differ in flight height (n = 17 males, mean ± SD: 2.7 ± 1.9 m; n = 17 females, 2.6 ± 2.0 m, p =.93). On average, males and females were followed for the same duration (mean ± SD for n = 20 males: 195 ± 96 s; n = 19 females: 243 ± 108 s; p =.156), and the estimate of airspeed was not related to measurement duration (p =.975).
With no tailwind, the airspeed of the males (y-intercept and SE: 4.24 ± 0.12) did not differ significantly from that of the females (4.04 ± 0.16). The fact that the mean airspeeds for male and female Phoebis sennae tended to differ (n = 20 males, mean airspeed ± SE, 4.23 ± 0.12 m/s; n = 19 females, 3.82 ± 0.19 m/s, p =.067) prior to adjusting for tailwind velocity underscores the importance of analyzing airspeed data in relation to ambient winds.
Both sexes significantly shifted their headings downwind over the course of
the day as wind speed increased (males, p =.043; females, p
=.0001), although females did so significantly more than males (one-tailed
Mann-Whitney test for angular differences: U = 240, p
<.05; Zar, 1999).
Furthermore, females turned downwind earlier than males (1300 h vs. 1400 h;
Figure 5), although the
difference is dependent on a small afternoon sample of males. Nevertheless,
tailwinds increased significantly over the course of the day for females
(n = 19, p =.0002), but only tended to increase for males
(n = 20, p =.051). Airspeed for females also declined over
the course of the day (n = 19, p =.012), but not for males
(n = 20, p =.478). Apparently, the females are turning
downwind, which will take them back to shore earlier than males.
|
On average, the butterflies track directions orient southwest (219° ± 22°). For much of the day, track directions changed counter to the solar azimuth (Figure 6). Track directions to the south and east were frequently associated with wind speeds exceeding 3.5 m/s (Figure 6). However, some butterflies that were not flying in high winds during the afternoon also flew southeasterly toward shore (Figure 6). Hence, the butterflies' orientation changed from westerly, out to sea, in the morning to southeasterly, toward land, in the afternoon, resulting in a southwesterly progressing, zigzagged route along the Atlantic coast of Colombia.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The southwesterly track direction of the migrating cloudless sulfur Phoebis sennae is directed toward the wet Atlantic forest of Darien, Panama, where they may spend the dry season in shadier, more humid conditions (ca. 180 km from the dry forest near Cartagena, Colombia; Figure 2). If the behavior of each individual that is sampled at a specific time indicates the behavior of the population over the course of the day, then a reasonable picture of the migration emerges. For those butterflies that are able to fly west far enough from shore, the strength and southwesterly direction of the trade wind prevalent offshore at this time of year (9-11 m/s; Clarke 1989
) may facilitate
arrival at the presumed destination site (a butterfly's ground speed would be
ca. 50 km/h). However, for butterflies that leave the coast too late or
otherwise fail to reach the region where trade winds predominate offshore,
southeasterly onshore winds increase in strength during the afternoon and
drift them toward coastal Colombia. These butterflies turn downwind toward a
coastal site that is nearer to Panama than the site at which they departed the
coast, and presumably continue to migrate southwesterly toward Panama on
subsequent days (see also Srygley,
2001).
In a tailwind, female P. sennae reduced their airspeeds, whereas
males did not. To the best of my knowledge, a difference among the sexes in
tailwind drift compensation has not been reported in insects. Females may have
reduced their velocity to a speed that maximizes range in the tailwind
(Pennycuick, 1978), such that
they minimize the consumption of lipids that would otherwise be allocated to
eggs and can safely reach the shore. In contrast, males may be selected to
minimize their flight time to the destination site by adopting a maximum
sustainable velocity. The males probably adopt Vmax rather
than Vmt because the flight is over water where there are
no stopover sites to replenish energetic resources.
By adopting a maximum sustainable velocity, a male's early arrival at the
breeding site may result in an increase in the number of copulations relative
to males that arrive later (also see
Myers, 1981). Protandry, or
the early emergence and entrance to the population by adult males, occurs in a
number of butterfly species (Nylin et al.,
1993; Wiklund and
Fagerström, 1977), and it is
strongly selected for in populations with discrete generations
(Singer, 1982). I suggest that
early arrival to suitable habitats by adult male butterflies is also selected
for in seasonally migrating populations.
It is reasonable to assume that female Phoebis sennae are sexually
immature; in other words, they exhibit the oogenesis flight syndrome typical
of other migrating insects (Johnson,
1969; Rankin and Burchsted,
1992). In Florida, the majority of migrating female P.
sennae were unmated (Walker,
1978). The steady immigration or local emergence of virgin females
would increase the mating success of early- relative to late-arriving males.
There may also be selection for males to emerge and migrate earlier than
females, although early migrants may lose safety in numbers that results from
migrating en masse. Nevertheless, flight speeds near Vmax
would result in the male's arrival ahead of other males and of females in its
consort that began migrating on the same day. The greater energetic cost of
flying at Vmax serves as a counterselective force.
Liechti et al. (1994)
developed a model for wind compensation that incorporates both crosswinds and
tailwinds. However, the model depends on compensation for crosswind drift, at
least in part. Four of 17 Phoebis sennae tested compensated for
crosswind drift over the Caribbean Sea
(Srygley, 2001). Therefore, it
is not possible to test a more complex model optimizing compensation for
crosswind and tailwind drift simultaneously with this data set.
To gauge the speed of a tailwind when flying over the sea, butterflies may use the sea surface as a ground reference. The accuracy of this method is compromised because the sea surface also moves in the direction of the wind, although not as quickly. A butterfly might also turn upwind and fly at a speed that holds landmarks on the horizon stationary, thus gauging the windspeed against its own airspeed. In this study, a landmark-based method would not have been possible in the 21 cases for which none was visible. Butterflies are occasionally, but rarely, observed circling when over water, which might provide useful information on ambient winds. However, behaviors that might provide insight into how butterflies gauge wind speed were not consistent.
|
ACKNOWLEDGEMENTS |
|---|
I thank C. Londono for use of his property on Isla Paloma in Islas San Bernardo as a base for these studies. I also thank R. Franke and S. Zea for their notes on migrants at Santa Marta, Colombia. I am grateful to A. Trimble for customizing his program to capture National Marine Electronics Association (NMEA) data, and to J. McNeil for comments on the manuscript. Financial support was generously granted by the National Geographic Committee for Research and Exploration. BBSRC Grant 43/FO8664 to A. Thomas supported the manuscript preparation.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Alerstam T, Hedenström A, 1998. The development of bird migration theory. J Avian Biol 29: 343-369.
Alerstam T, Lindström A, 1990. Optimal bird migration: the relative importance of time, energy, and safety. In: Bird migration: the physiology and ecophysiology (Gwinner B, ed). Berlin: Springer; 331-351.
Brown FM, Heineman B, 1972. Jamaica and its butterflies. New York: E.W. Classey.
Clarke J, 1989. Atlantic pilot atlas. London: Stanford Maritime.
Dudley R, DeVries PJ, 1990. Flight physiology of migrating Urania fulgens (Uraniidae) moths: kinematics and aerodynamics of natural free flight. J Comp Physiol A 167: 145-154.
Heran H, Lindauer M, 1963. Windkompensation und Seitenwindkorrektur der Bienen beim Flug über Wasser. Z vergl Physiol 47: 39-55.
Johnson CG, 1969. Migration and dispersal of insects by flight. London: Methuen.
Liechti F, Hedenström A, Alerstam T, 1994. Effects of sidewinds on optimal flight speed of birds. J Theor Biol 170: 219-225.
May PG, 1992. Flower selection and the dynamics of lipid reserves in two nectarivorous butterflies. Ecology 73: 2181-2191.[Web of Science]
Myers JP, 1981. A test of three hypotheses for latitudinal segregation of the sexes in wintering birds. Can J Zool 59: 1527-1534.
Nylin S, Wiklund C, Wickman P-O, Garcia-Barros E, 1993. Absence of trade-offs between sexual size dimorphism and early male emergence in a butterfly. Ecology 74: 1414-1427.[Web of Science]
Oliveira EG, Srygley RB, Dudley R, 1998. Do Neotropical migrant butterflies navigate using a solar compass? J Exp Biol 201: 3317-3331.[Abstract]
Pennycuick CJ, 1978. Fifteen testable predictions about bird flight. Oikos 30: 165-176.[Web of Science]
Pivnick KA, McNeil JN, 1987. Puddling in butterflies: sodium affects reproductive success in Thymelicus lineola. Phys Entomol 12: 461-472.
Rankin MA, Burchsted JCA, 1992. The cost of migration in insects. Annu Rev Entomol 37: 533-559.[Web of Science]
Richardson WJ, 1991. Wind and orientation of migrating birds: a review. In: Orientation in birds (Berthold P, ed). Basel: Birkhäuser Verlag; 226-249.
Singer MC, 1982. Sexual selection for small size in male butterflies. Am Nat 119: 440-443.[Web of Science]
Smedley SR, Eisner T, 1996. Sodium: a male moth's gift
to its offspring. Proc Natl Acad Sci USA
93: 809-813.
Srygley RB, 2001. Compensation for fluctuations in crosswind drift without stationary landmarks in butterflies migrating over seas. Anim Behav 69: 191-203.
Srygley RB, Oliveira EG, 2001. Migration patterns and orientation mechanisms within the flight boundary layer. In: Insect movement: mechanisms and consequences (Reynolds D, Thomas C, Woiwod I, eds). London: CAB International; 183-206.
Srygley RB, Oliveira EG, Dudley R, 1996. Wind drift
compensation, flyways, and conservation of diurnal, migrant neotropical
Lepidoptera. Proc R Soc Lond B 263:
1351-1357.
Walker TJ, 1978. Migration and re-migration of butterflies through north peninsular Florida: quantification with Malaise traps. J Lepidopt Soc 32: 178-190.
Walker TJ, 1991. Butterfly migration from and to peninsular Florida. Ecol Entomol 16: 241-252.
Walker TJ, Littell RC, 1994. Orientation of fall-migrating butterflies in North peninsular Florida and source areas. Ethology 98: 60-84.[Web of Science]
Wiklund C, Fagerström T, 1977. Why do males emerge before females? A hypothesis to explain the incidence of protandry in butterflies. Oecologia 31: 153-158.[Web of Science]
Zar JH, 1999. Biostatistical analysis. Upper Saddle River, New Jersey: Prentice Hall.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. B. Srygley and R. Dudley Optimal strategies for insects migrating in the flight boundary layer: mechanisms and consequences Integr. Comp. Biol., July 1, 2008; 48(1): 119 - 133. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||