5
TOWARDS PREDICTIVE MODELS OF BIRD
MIGRATION SCHEDULES:
THEORETICAL AND EMPIRICAL BOTTLENECKS
Bruno J. Ens, Theunis Piersma and Joost M. Tinbergen
DECISIONS
REFUEL?
STOP
L
MIGRATION ?
DEPART ?
v / :
GOAL
■ REPRODUCE
PREPARE FOfiV ■ S U R V IV E
MIGRATION ?J
PREY
ECOLOGY
Nederlands Instituut voor Onderzoek der Zee
© 1994
This report is not to be cited without
acknowledgement of the source.
Netherlands Institute for Sea Research (NIOZ)
P.O. Box 59, 1790 AB Den Burg,
Texel, The Netherlands
Netherlands institute for Forestry and Nature Research
(IBN-DLO), P.O. Box 167, 1790 AD Den Burg, Texel
The Netherlands
Behavioural Biology, The Zoological Laboratory
University of Groningen, P.O. Box 14, 9750 AA Haren
The Netherlands
ISSN 0923-3210
Cover design: H. Hobbelink
TOWARDS PREDICTIVE MODELS OF BIRD
MIGRATION SCHEDULES:
THEORETICAL AND EMPIRICAL BOTTLENECKS
Bruno J. Ens1, Theunis Piersma2 and Joost M. Tinbergen3
Netherlands Institute for Forestry and Nature Research (IBN-DLO),
P.O. Box 167,1790 AD Den Burg (Texel), The Netherlands
N etherlands Institute for Sea Research (NIOZ), RO. Box 59,
1790 AB Den Bureg (Texel), The Netherlands
3Behavioural Biology, The Zoological Laboratory, University of Groningen,
P.O. Box 14, 9750 AA Haren, The Netherlands
Report of a w orkshop made possible through grants from the Dutch National Research
P rogram m e on G lobal A ir and Clim ate Change (NOP), the Institute for Forestry
and Nature Research (IBN-DLO) and the Netherlands Institute for Sea Research (NIOZ)
NETHERLANDS INSTITUTE FOR SEA RESEARCH
NIOZ-RAPPORT 1994 -5
A few words on how this meeting came about and what made it possible may be of interest. The idea for the
meeting arose during a visit to Oxford. The combined presence of Peter Hicklin and BE prompted Alasdair
Houston to organize a small meeting to discuss models of migration. This impromptu meeting was also
attended by Richard Caldow, John McNamara, Tony Moodey, Bill Sutherland, Thomas Weber and Wolfgang
Weisser. In part because the meeting worked so well, it seemed that more people could have profited from
attending. Also, the discussions might have been even more lively with a somewhat greater audience and a
greater share of die-hard empiricists, willing to harp on every possible weakness in the models. Finally, there
was a feeling that the study of migration strategies was rapidly gaining momentum and that this provided an
important window of opportunity to forge a link between the empirical and theoretical studies of migration and
thereby set them on a common course. Phoning around made it clear that many people were happy to attend
such a “theoreticians-meet-empiricists" meeting. Initially envisaged as a small meeting of at maximum 35 sci­
entists it grew to a meeting of almost double that size through word of mouth in combination with a deliberate
policy to let enthusiasm have its way. Wim Wolff showed great enthusiasm at the first mention of the idea and
helped to organize funds. These funds were used to pay for the travel expenses of some dearly wanted partic­
ipants, mainly from the other side of the Atlantic, and for the publication of this report. Without the willingness
of the majority of the participants to pay for their own expenses, the meeting would not have been possible
though. Funds were obtained from the National Research Programme on Global Air and Climate Change, the
Institute for Forestry and Nature Research and the Netherlands Institute for Sea Research. The cooperative
attitude of Joke Hart, Michaela Scholl and Gjalt Steenhuizen during the meeting substantially contributed to its
success. We are also grateful for the many replies that we received to our request for comments on an earlier
version of this report. Comments were received from: Thomas Alerstam, Herbert Biebach, Bruno Bruderer,
Dan Cristol, Peter Evans, John Goss-Custard, Anders Hedenström, Richard Holmes, Alasdair Houston,
Lukas Jenni, Susi Jenni-Eiermann, Marcel Klaassen, Felix Liechti, Äke Lindström, Jaap van der Meer, Frank
Moore, Martin Morton, Berry Pinshow and Thomas Weber. Especially gratifying were the comments of Ellen
Ketterson, a participating non-participant, and the long letters of Paul Dolman, full of healthy disagreement.
Henk Hobbelink enthusiastically engineered a cover for this report. The report was produced by Anneke Bol
and Nelleke Krijgsman.
TOWARDS PREDICTIVE MODELS OF BIRD MIGRATION SCHEDULES:
THEORETICAL AND EMPIRICAL BOTTLENECKS
Bruno J. Ens, T h e u n is P ie rsm a & Jo o st M. T inb erg en
"The relation of theory to experiment in biology has been an uneasy one. The
word “theoretical" has generally had perjorative connotations, and the right to the­
orize was the reward for years o f laboratory and field work. In fields where
progress depends mostly on the refinement of technique in order to facilitate more
accurate description, this perhaps did not matter too much, but other areas suf­
fered from an indigestion o f facts, while data was collected without reference to
problems. In these circumstances, theoretical work often diverged too far from life
and became exercises in mathematics inspired by biology rather than an analysis
o f living systems. “
(L e v in s , 1968)
INTRODUCTION
T ake the a rch etypa l w ader, b re ed ing on th e arctic
tundra, using in te rtida l m u d fla ts in the te m p e ra te
zone as a s to p o v e r site during m ig ration and w in te r­
ing on in te rtida l m ud flats in the tropics. W h a t w ill h a p ­
pen to the birds if so m e o f th e ir h a bita ts are m odified
by m an fo r exp lo itation , like reclam atio n o f the in te r­
tidal m u d fla ts o r oil-d rillin g on th e tu n d ra ? D esired
a n sw e rs w o u ld a lm o st ce rta in ly in clud e exp ected
cha ng es in th e nu m b e r and tim in g o f birds using the
affected areas, as w ell as ch a n g e s in th e ir pattern of
m ass gain. It w ill n o t be ea sy to g ive such answ ers. It
w ill be even m ore d ifficu lt to p re d ict w h a t w ill happen
if all h a bita ts are s im u lta n e o u sly affe cte d by global c li­
m ate change. O ne w a y to go a b o u t an sw e rin g the se
q u e stio n s is to tre a t the m as th o u g h t exp e rim e n ts
w h ere o u r a b ility to co m e up w ith an a n sw e r in dica tes
the cu rre n t s ta te o f o u r kno w le dg e. So, w h a t is the
cu rre n t state o f o u r kno w le dg e?
T h e re is no d o u b t th a t stu d ie s o f the m ig ratory
m ove m e n ts o f birds have a long history. A ltho ug h
ea rly stu d e n ts o f m ig ra tio n did n o t fail to th e o rize (see
review in G a u t h r e a u x 1982), a tte m p ts to co n stru ct
form al m a th e m a tica l m odels, a n e ce ssa ry p re re q u i­
site fo r pre cise pre dictions, are o f m uch m ore rece nt
origin. In the late sixtie s and e a rly seve ntie s P e n n y c u ic k (1969) and T u c k e r (1973) p re sen te d basic
flig h t m e ch a n ica l th e o rie s a b o u t bird m ig ration . In the
s e ve n tie s fo rm a tio n flig h t and w ind effects w e re e v a l­
uated (P e n n y c u ic k 1978, A ler s t a m 1979). In the
eig h tie s ga m e th e o ry (see M ay n a r d S m it h 1982) was
used to stud y the co n d itio n s fo r pa rtial bird m igration
(L u n d b e r g 1987) and sta b le g e o g ra p h ica l po pulation
s e g re g a tio n (L u n d b e r g & A le r s t a m 1986). In the
nine ties s im p le o p tim iza tio n p ro ce d u re s w ere used to
pre d ict op tim al d e p a rtu re fa t lo ad s (A lerstam & L in d ­
s t r ö m 1990). In ad dition in cre a sin g use is being
m ad e o f sto ch a stic d yn a m ic pro g ra m m in g to handle
m ore co m p le x d yn am ical a sp e cts o f m igration. The
u se fuln ess o f th is m a th e m a tica l te ch n iq u e in the
stu d y o f fo ra g in g and life histo ry pro blem s w as
bro u g h t to the fo re by M c N a m a r a & H o u s t o n (1986)
and M a n g e l & C l a r k (1986). A s th e o rigin al flig ht
m ech an ical th e o rie s w e re a lso co n tin u a lly im p roved
and adde d upon th ro u g h o u t th is perio d (R a y n e r
1988, P e n n y c u ic k 1989), th e re can be no do ub t tha t
m athe m atica l m od els th a t a cco u n t fo r th e variou s
asp e cts o f m ig ration are d e ve lo p e d a t an in crea sing
rate.
A ltho ug h th is is a h e alth y d e ve lo p m e n t, th e re is a
risk th a t the th e o re ticia n s w ill b e co m e so excite d by
each o th e r th a t th e y d isa p p e a r en m asse in to o u te r
space, instead o f con fro ntin g th e m ore e m p irica lly
m inded stu d e n ts o f m ig ration w ith pe n e tra tin g q u e s ­
tion s. S uch qu e stio n s can be o f g re a t help in w a rdin g
o ff the e ve r-p re se n t d a n g e r o f d ro w n in g in th e c o m ­
plexitie s o f nature. In return, th e o re tic ia n s w ill profit
from rem aining a n cho red to th e real w o rld, even
th o u g h it m ay be p ro fo u n d ly irrita ting to be told the
fa cts o f life by m e m b e rs o f the scie n tific com m u nity
w ith m ud on th e ir boots. U sing m ore d e fe re n tia l la n ­
gu a g e it am o un ts to the c o n victio n th a t scie ntific
pro gre ss is ha m pe red w h en th e o re tica l a d va n ce s are
n o t a cco m pa nie d by c o m p a ra b le a d va n ce s on the
em p irica l sid e o f th e rese arch pro gra m . O n ly th ro u g h
close co o pe ration will it be po ssib le to a ch ie ve
ad van ces on both sides. A cle a r e xa m p le o f w h a t
m ay go w ro ng is niche theory, w h ich so u g h t to e xp la in
the pa tte rns in anim al co m m u n itie s as a re su lt o f
in te rsp e cific co m p e titio n (M a c A r t h u r 1972). W h ile
th e o ry be cam e in cre a sin g ly co m p le x, in su fficie n t
ad van ces w e re m ade in th e d e v e lo p m e n t o f a m e th ­
od o lo g y th a t allow ed the actu al m e a su re m e n t o f lim it­
ing reso urce s in the field. A s a result, th e rese arch
pro gra m w a s not proven w ro ng o r su b sta n tia lly m o d i­
fied on the ba sis o f em p irica l evide nce, b u t sim p ly
a b a n do ned by th e m a in stre a m o f e co lo gists.
It is th e re fo re to be ho pe d th a t the w o rksho p,
o rg an ized on Texel fro m 6 to 8 Ju ly 1993, su cce e d e d
in esta b lish in g firm links be tw ee n th e o re tic ia n s and
e m p iricists, th ro u g h the d ire ct co n fro n ta tio n o f ideas
w ith facts and vice versa, as w ell as th ro u g h the
e sta b lish m e n t o f pe rson al co n ta cts. D e spite the large
n u m b e r o f p a rticip ants the m ee ting retained its
in te nd ed inform al character. T hough discussio ns
w e re lively, p a rticip ants sp e n t th e ir energy in
exch a n g in g ideas instead o f d isp la y be ha viou r
in te nd ed to im press, o r cla im priority. T his sug ge sts
th a t clo se personal con tacts are a rem e dy a g a in st the
trend noted by M a d d o x (1993) th a t in te nsifying social
co m p e titio n am o ng scie n tists is in cre a sin g ly crip pling
scie nce th ro u g h im p ed ing the fre e flo w o f ideas. The
re m e d y relies on th e o b serva tion th a t scientists, like
o th e r anim als, ca n n o t esca pe being p riso n e rs o f th e ir
social relatio nships. S in ce lim its to th e tim e b u d g e t o f
an anim al w ill lim it th e tim e a va ila ble fo r social c o n ­
tacts, the nu m b e r o f su fficie n tly in tim ate relatio nships
w ill be lim ite d too, su g g e stin g th a t the rem e dy does
not w o rk w h e n the n u m b e r o f scie ntists is too large.
DEFINING THE PROBLEM
Birds are on ly ca p a b le o f com p le ting a full m ig ratory
cycle if th e y possess fo rm id a b le p o w e rs o f n a vig a ­
tion. No w o n d e r th a t m uch rese arch e ffort has co n ­
cen tra ted on trying to un de rsta nd th e m ech an ism s o f
na vig ation (Papi & W a l r a f f 1982). S im ilarly, s u c ­
cessful com p le tion o f the m ig ra to ry cycle req uire s a
pro pe r tim in g. M ainly throu gh the w o rk o f G w in n e r
(e.g. 1990) e xp e rim e n ts w ith m ig ra to ry birds have
provided so m e o f the be st de m o n stra tio n s o f the
existe nce o f e n d o g e n o u s circa nn ual rhythm s s y n ­
chro nized by en viro n m e n ta l Z e itg e b e r s . H ow ever, an
u n de rsta nd ing o f som e o f the abilities th a t a llo w birds
to m ake th e ir m ig ra to ry jo u rn e ys is a fa r cry from a
th e o ry pre dicting th e de tails o f a m ig ration sche du le
o f a p a rticu la r bird spe cie s (P ie rsm a 1987, A le r s ta m
& L in d s trö m 1990, Ens et al. 1990).
B a k e r (1978) fo rm u la te d a ve ry general m odel and
cou che d it in such a b stra ct term s th a t it rem ains
"u nfam ilia r to m o st11acco rding to K e t t e r s o n & N olan
(1983). T hese au thors s u b se q u e n tly state th a t “w h ile
the co m p re h e n s ive n e ss of the m odel m akes it a d m i­
rable in the abstract, in practice it m ay ren de r p re dic­
tion s un testable". T h e alte rn a tive to con stru cting a
ve ry general and co m p re h e n sive th e o ry e n co m p a ss­
ing all possible typ e s o f m ove m en t in the en tire a n i­
m al kin gd om is to build m odels fo r w e ll-d e fin e d and
th e re fo re n e ce ssa rily lim ited problem s. T his w as
m ore o r less the ap p ro a ch im p licitly o r e xp licitly take n
by the p a rticip ants o f the w o rksho p. Like B a k e r
(1978) it w as g e n e ra lly a ssu m ed tho ugh th a t the
de tails o f a m ig ration sch e d u le are sha pe d by natural
selection.
T he fa c t th a t a t p re se n t none o f the pa rticip ants
see m ed to be co n stru ctin g a grand un ifyin g the ory
should n o t be take n to m ean th a t such a th e o ry is not
w anted. In con tra st, an im p o rta n t goal o f the w o rk ­
shop w as to id en tify how the va rio u s studies o f s p e ­
cific m ig ratory d e cisio n s fit together. A fter all, the re
can be no do ub t th a t so lu tio n s to o n e pro blem c o n ­
strain the po ssib le so lu tio n s to o th e r pro b le m s. To
avo id m isu n d e rsta n d in g s it is im p o rta n t to m a ke cle ar
th a t in th is re p o rt the w ord d e cisio n is used in the
se n se o f M c F a r l a n d (1977) fo r case s w h e re seve ral
m utua lly e xclu sive b e ha viou ral o p tio n s e x is t fo r the
anim al: "the re m u st be so m e pro cess o r m ech an ism
th a t d e te rm in e s w h ich activity is to ha ve prio rity at
any p a rticu la r tim e, and such a p ro ce ss is c o n v e n ­
ie ntly term ed a de cisio n pro cess". T hus, the use o f
the w ord de cisio n is in no w a y in te nd ed to im p ly c o n ­
scio us ch o ice by th e anim al. R g u re 1 d e p icts an id e ­
alized de cisio n sch e m e o f a m ig ra to ry bird. The
sim p licity o f th e sch e m e e ffe ctive ly h id e s the m an y
discu ssio n s be tw ee n the th re e o f us d u rin g its g e s ta ­
tion. A cco rd in g to the sch e m e the m ig ra to ry jo u rn e y
o f an in dividu al bird can be broken do w n into fo u r
m ajor d e cisio ns: (1) the de cisio n to s ta rt p re p a ra tio n s
ne cessa ry fo r m ig ration like building up fu e l stores,
(2) th e de cisio n to d e p a rt on the m ig ra to ry flig ht, (3)
the de cisio n to in te rru p t th e m ig ra to ry flig h t fo r re fu e l­
ling and (4) the de cisio n to end m ig ration a lto g e th e r
and sea rch fo r a good site to bre ed o r su rvive the
no n -b re e d in g season. T h e se d e cisio n s lead to on e o f
three phases: the fue llin g phase, th e flig h t ph a se and
the re sid e n t phase. D uring each phase co n tin u o u sly
a m ultitu de o f d e cisio n s has to be m ad e in resp on se
to the pre vailin g co n d itio n s in o rd e r to reach the
ap p ro p ria te state ch a ra cte rizin g th e phase. During
the fu e llin g and refue lling pe rio ds the a n im a l has to
d e cid e on the rate o f m ass gain in re sp o n se to local
foo d co n d itio n s and pre da tion risks. D u ring the m ig ra ­
to ry flig h t the an im a l has to d e cid e on th e po sitio n in
th e flock, the flig h t spe ed and th e flig h t a ltitu d e in
resp on se to the w e a th e r co n d itio n s en route. Finally,
the bird has to settle dow n, ta kin g h a b ita t q u a lity and
th e p re se n ce o r ab sen ce o f c o m p e tito rs in to account.
T he sch e m e has bro ken dow n the pro blem in to a
set o f decisions, allow in g an e n q u iry into th e costs
Major m&E^jDRy
DEÔ sioM S
s
PREY -
J
ecoLc&y ~
*
Fig. 1. Idealized scheme of the major migratory decisions
faced by a migrating bird (see text for a detailed discussion).
and be n e fits (w hich m ust u ltim a te ly be e xp resse d in
te rm s o f fitne ss), as w e ll as th e u n de rlyin g m e ch a ­
nism o f each decision. A t th e sam e tim e the sche m e
high lig hts th e in te rd e p e n d e n ce o f the de cisio ns:
tho ugh th e va rio u s de cisio n s m a y be stud ied as in d e ­
p e nd en t p ro b le m s in th e ir ow n right, th e final solution
req uire s th a t th e resu lts o f th e se stu d ie s fit in the
sam e fram e. W h a t the s c h e m e do es n o t d e p ict is the
d ire ctio n a l co m p o n e n t o f som e o f th e de cisio ns. A p a rt
from d e cid in g w h a t to do w hen, th e anim al m ust also
de cid e w h e re to go. F urthe rm o re, th o u g h the sche m e
pre sen ts the problem as one o f the in dividu al bird fa c ­
ing a seq ue nce o f b e ha viou ral decisions, refere nce to
a lo w er level o f in te gra tion, i.e. p h ysio log ica l p ro c­
e sse s in side the anim al, an d a h ig h e r level o f in te g ra ­
tion, i.e. po pu latio n pro cesse s, w ill be necessary.
P h ysio lo g y fo r instance, d e te rm in e s how fa r a bird
can fly w ith a given fa t load a t a given w ind speed,
w h ile the costs and be n e fits in te rm s o f fitne ss o f a
pa rticu la r m ig ra to ry de cisio n w ill often depend on the
m ig ra to ry b e h a vio u r o f co n sp e cifics, i.e. on p o p u la ­
tion p ro cesse s. In fact, in ste ad o f de scrib ing the
m ig ration sc h e d u le as being ad h e re d to by a h yp o ­
th e tica l sin g le individu al m an y o f us u ltim a te ly w a n t to
in clud e the n u m b e r o f in dividu als in the de scrip tion o f
a m ig ration schedule.
D
weight
Power,
P(watt)
THE MIGRATORY FLIGHT
FLIG H T M ECHANICS
F lig h t m e ch a n ics is a ba sic e le m e n t in th e o rie s o f bird
m ig ration (A le rs ta m 1991). A cen tra l e le m e n t of
flig h t m e ch a n ics is the p o w e r curve, w h ich predicts
th a t e n e rg y e xp e n d itu re w ill be high a t both low and
high flig h t s p e e d s and reach a m in im um at in te rm e d i­
ate sp e e d s (P e n n y c u ic k 1989). T h is p o w e r curve can
be use d to fin d th e flig h t spe ed th a t m in im ize s the
e n e rg y co st per u n it tim e, th e flig h t spe ed fo r m in im iz ­
ing th e e n e rg y co sts p e r u n it o f d ista n ce covered (the
m a xim u m ra n g e speed) and th e op tim al flig h t speed
fo r m in im izing th e total d u ra tio n o f th e m ig ra to ry jo u r­
n e y (Fig. 2). A le r s ta m e l at. (1993) o b serve d fo r a
range o f s e a b ird s th a t a ve ra g e a irsp e e d s fell betw een
th e m in im um po w e r and m a xim u m ran ge sp e e d s e s ti­
m a te d from a e ro d yn a m ica l theory. B r u d e r e r &
W e itn a u e r (1972) sh o w e d th a t C o m m on S w ifts
rem a in a irb o rn e du rin g s le e p in g flig h ts a t a b o u t the
th e o re tica l spe ed o f m in im um po w e r and clo sely
Rate of fuel deposition,
K (watt)
Fig. 2. Power curve estimated for the Arctic Tern according to the theory of bird flight mechanics. The mechanical power
output indicated on the power axis should be divided by the conversion efficiency (normally assumed to be about 23%) to
give the total chemical energy metabolized by a flying bird. The rate of fuel deposition is converted into its mechanical equiv­
alents to be compatible with the flight power given. V mp is the speed that minimizes energy costs per unit of time. Vmr is the
speed that minimizes the energy costs per unit of distance covered. V , is the optimal flight speed for minimizing the total
duration of a migratory journey, including the periods of fuel deposition. Vmjqr is the maximal overall speed of migration that
can be attained. From A lerstam (1991). Reprinted with permission from Elsevier Tredn Journals.
Power:
r max
Speed:
V = f (Vz )
Time:
*1
w
to ”
2
x2
(V mr + W )
E nergy:
Fig. 3. Schem atic migration flight with an initial clim bing phase to altitude = z w hen distance
is covered and a cruising
phase when distance x , is covered. W e have indicated the qualities needed to derive the total energy cost (E, + E2) for the
entire migration flight, which is the currency being m inimized with respect to clim b rate (see H e d e n strö m & A le rs ta m , 1994
for details)
match maximum range speed during migratory
flights. L ie c h t i (1992) showed that small passerines
adjust their airspeed to wind conditions in close
agreement with the theoretical curve.
Anders Hedenström applied the power curve to cal­
culations on climbing flight. A simple model predicting
the optimal rate of climb for birds setting out on a
migratory flight was presented (Fig. 3). The model
uses an energy minimization argument for the entire
flight (climbing and cruising phases) and gives the
climb rate resulting in minimum cost of transport.
Some data in support of the model was presented of
shorebirds departing from NW Iceland during spring
migration (H e d e n s t r ö m & A l e r s t a m 1994). Compar­
ing species, a negative relation between climb rate
and body mass was predicted, and observed. A neg­
ative correlation between maximum fuel load capac­
ity, i.e. the fuel load at which climbing flight is still
possible, and body mass was also predicted and
observed. As the maximal flight range is determined
by the maximal fuel load capacity it follows that
potential flight ranges are size-dependent (HEDEN­
STRÖM & A l e r s t a m 1992).
Bruno Bruderer (in press) reported on the first
study combining on-site radio sonde measurements
with reliable radar observations on height distribution
of migrants crossing the Negev desert in Israel. In
addition, the data were gathered under extreme tem­
perature and humidity conditions, and by this, ena­
bled Bruderer and co-workers to test existing
hypotheses on the influence of temperature and
humidity on altitude distribution. A statistical model,
developed in cooperation with Les Underhill (B r u d ­
e r e r et al. in press), indicated that choice of height
stratum was primarily affected by the amount of tail
wind, contradicting suggestions (see later) that noc­
turnal migrants above desert areas have to adjust
their flight level to minimize evaporation of water. The
model, which assumes that the birds compare the
meteorological conditions in the adjoining height
strata and then switch to the more profitable layer
with a certain estimated probability, yielded a close fit
to the empirical data: the change in tailwind speed
predicted on average 63% of the autumn and 56% of
the spring distributions, while none of the other varia­
bles had a significant influence. Fig. 4 provides an
example.
Felix Liechti combined the model of P e n n y c u ic k
(1969, 1989) for maximum range speed with A l e r s t a m 's (1979) model on optimal drift strategy, calcu­
lating the optimal adjustment of heading and airspeed
to compensate for changeable winds with respect to a
specific goal (L ie c h t i 1992, 1993 and in press). In this
case flying the maximum range is regarded optimal
(Fig. 5). Under most wind conditions the proposed
statistical model on optimal behaviour came close to
the observed behaviour, but large deviations from the
model occurred under strong side winds (above 5 m/
s).
E N E R G Y A N D O T H E R P O T E N T IA L L Y L IM IT IN G
R E S O U R C E S D U R IN G M IG R A T IO N
Under the assumption that only energy is critical,
maximal flight range can be calculated for a given fat
load from the previously discussed power curve.
However, the assumption may be wrong. There may
be conditions (i.e. ambient temperatures) where
14./ 15.9.91
km
4
Fig. 5. Optimal flight behaviour: m inim izing energy exp e nd i­
ture during one flight stage in relation to a specific goal. The
distance flown in one stage (At.Vg) is the sum of the work
done by the bird (At.Va) and the displacem ent by the wind
(At.Vw). For an optimal behaviour the rem aining distance
after the flight stage (X) m ust be minimal. (S1 ^starting point,
S2=stopover site, G=goal, D=initial distance to the goal,
ra=wind direction, ß=direction of heading, Va=airspeed,
Vw=wind speed, Vg=groundspeed, At=flight tim e). From
LlECHTl (1992).
□
water loss, instead of energy expenditure, limits flight
range (Y a p p 1 9 6 2 , B ie s e l & N a c h t ig a l l 1 9 8 7 ). In a
recent model C a r m i et al. (1 9 9 2 ) have sought to
quantify these conditions. Marcel Klaassen (in press)
subsequently refined the model of Carmi et al. with
respect to the feedback of the difference between
metabolic water production and total water loss (i.e.
net water loss) on the flight costs. The flow chart of
the modified model is presented in Fig. 6.
ENERGY
O
WATER
0 5 10 15 20 25 30 35
“ I
I
I
I
I
I-----------1
10 birds/km3
Fig. 4. C om parison of predicted (shaded bars) and observed
(white bars) height distribution of nocturnal bird m igration in
autum n (14/15.9.1991) over the Arava Valley (Israel). Pre­
dictions are based on the difference in tailwind between
neighbouring height levels. The circular diagram in the
upper right corner indicates the wind direction and speed
(length o f the bar) for three height intervals (inner circle 0 to
1000; m edium 1001 to 2000; outer 2001 to 3000 m above
ground level) and throughout the night (white 20:00 h, black
24:00 h, shaded 4:00 h). From B r u d e r e r et a i (in press).
Fig. 6. R ow chart o f the physiological processes underlying
the com puter model of Carmi et at. (1992) and including the
refinem ent of K la a sse n (in press). Since the bird loses
mass continuously during the flight the program perform s
iterations over sm all tim e steps. The variables affected by
the refinem ent of KlAASSEN (in press) are indicated with a *.
When meteorological conditions are known, the
water loss model, which necessarily includes a calcu­
lation of the energy balance, allows the prediction of
the flight altitude that maximizes flight range. Radar
observations by Biebach and Klaassen on migrants
crossing the Egyptian desert showed that observed
flight ranges corresponded more closely to the pre­
dictions from the water loss model than to the predic­
tions from the energy balance model alone. The
water loss model also predicts that flight range can be
increased by interrupting the trans-Sahara flight dur­
ing the day. Analysis of radar observations on the pat­
terning of the passage of migrants at different
locations by Herbert Biebach indicates that many
migrants do indeed rest during the day, as suggested
by B ie b a c h et al. (1991). Further support for such
interruptions was provided by the observations with
tracking radar of B r u d e r e r et al. (in press). However,
since in Bruderer's data on migrants crossing the
Negev desert the birds did not adjust their flight levels
according to the temperature or humidity profile, they
did not seem to be limited by water loss (B r u d e r e r et
al. in press).
Though the water loss model has met some empiri­
cal success it was generally agreed that many impor­
tant parameters and processes were poorly known.
The following two questions were identified as the
most burning: (1) how much water can a bird lose
before suffering physiological damage, (2) what is the
rate of water loss under various conditions? Given
that, apart from energy, the water balance may some­
times be crucial in determining flight range, it was felt
that it might pay to take an even broader perspective
and also consider the role of protein turnover and the
need for sleep. In fact, there was a general consen­
sus that fat, protein and water constituted a complex
of components potentially limiting the migratory flight.
THE MIGRATORY JOURNEY: (RE)FUELLING,
DEPARTING AND FLYING
In the above we have discussed decisions that need
to be made during the flying phase. In the following
we will discuss decisions and physiological adapta­
tions during the fuelling phase alone, or in combina­
tion with the flying phase.
P H Y S IO L O G Y O F M A K IN G A N D U S IN G N U T R IE N T
STORES
glycerol
FFA bound
to albumin
Fig. 7. Sim plified model of fatty acid transport from adipose
tissues to the muscles. Black arrows indicate the usual path­
way. Double-lined arrows indicate the pathway, which is
suggested to help to increase fatty acid delivery to the flight
muscles during flight in small migrants. FFA = free fatty
acids; TG = triglycerides; VLD L = very low density lipopro­
teins. Based on Jenni-E ierm ann & Jenni (1992).
Lukas Jenni reported on his physiological studies on
migrant passerines caught when crossing the Alps
(J e n n i -E ie r m a n n & J e n n i 1991, J e n n i & J e n n i -E ie r ­
m a n n 1992). Measurement of the plasma level of trig­
lycerides (the form in which fat is normally
transported to the adipose tissue) suggests that small
passerines resynthesize triglycerides in order to
achieve a high rate of fat utilization during migration.
This hints at an additional metabolic pathway not nor­
mally used by mammals during exercise (J e n n i -E ie r ­
mann
& J e n n i 1992; Fig. 7). Furthermore, day
migrants seemed to rely less on fat compared to night
migrants. Thus, it appears that differences in meta­
bolic physiology are closely linked to the migratory
strategy of a species.
S u s i J e n n i - E ie r m a n n & J e n n i (in press) showed
that plasma triglyceride levels in the blood of birds
during stopover can be used to predict the amount of
mass change, i.e. the amount of fat stored, during
that day. This allowed them to conclude on the basis
of measurements on birds caught only once that
insectivorous birds store less fat during days when
feeding conditions seem poor, i.e. after rainfall and at
high wind speeds (J e n n i & J e n n i -E ie r m a n n unpub­
lished). This biochemical technique may prove a pow­
erful tool for evaluating strategies of fuel storage in
the field as it circumvents many of the estimation
problems discussed by Z w a r t s et al. (1990).
C O N S IS T E N C Y IN A L L O M E T R IC R E L A T IO N S H IP S O F
P H Y S IO L O G IC A L A N D E C O L O G IC A L V A R IA B L E S
Jaap van der Meer, in collaboration with Theunis
Piersma, had embarked on an evaluation of various
allometric relationships between a dimension of body
size and variables like maximal intake rate, energy
expenditure during resting and foraging and energy
cost of thermoregulation. When these allometric rela­
tionships are formalized in a consistent manner, it can
be investigated in what way the shape of some
selected relationships constrain the shape of the
other relationships, assuming the animal retains
energy balance. It is also possible to enquire into the
conditions under which maximum fat deposition rate
shows a negative correlation with lean body mass, as
reported by L in d s t r ö m (1 9 9 1 ).
O P T IM A L D E P A R T U R E FAT LO A D S
All models that seek to explain departure fat loads as
optimal decisions necessarily include assumptions
about the flight phase. The marginal value, with
respect to flight distance, of adding an extra gram of
fuel will decrease with increasing fuel loads due to the
increased flight costs. From this the optimal departure
fat load can be calculated for a given fat deposition
rate and a given optimization criterion (Fig. 8, A l e r ­
s t a m & L in d s t r ö m 1 9 9 0 ). Possible criteria are mini­
mizing the time spent on the migratory journey, or
minimizing the energy spent on covering a given dis­
tance. Though the positive relationship between
departure fat load and fat deposition rate predicted by
the time minimization hypothesis was found in a field
experiment with Bluethroats, the observed departure
fat loads remained below the predicted levels (L in d STRÖM & A l e r s t a m 1 9 9 2 ). This could be due to not all
individuals having the same expected speed of
migration, or to variations in the expected speed of
migration along the migration route.
Assuming that increasing fat loads increase the risk
of predation, smaller departure fat loads are also
expected when birds attempt to maximize safety from
predation during the migratory journey (A l e r s t a m &
L in d s t r ö m 1 9 9 0 ). Experiments by Äke Lindström to
test if captive Chaffinches and Bramblings would put
on less fat when regularly confronted with a model
Sparrowhawk yielded mixed results.
Crucial for the theory developed by A l e r s t a m &
L in d s t r ö m (1 9 9 0 ) is a time or energy cost during a
settling phase preceding fat deposition in a stopover
site. During the discussions it was questioned if such
a phase really exists. The study of Frank Moore on a
barrier island (see M o o r e et al. 1 9 9 0 ) where
migrants are caught upon arrival and immediately fol­
lowed using telemetry, seemed ideally suited to
obtain the necessary insights on this settling phase.
The model of Alerstam and Lindström is most appli­
cable to birds that migrate over areas where suitable
Fuel load, f
Fig. 8. Prediction of optim al fuel loads, (a) The relationship
between flight range (Y) and fuel load (f) at departure will be
negatively accelerated because of increased flight costs
with increasing fuel loads, (b) As a consequence the m ar­
ginal rate of gain in potential flight distance (dY/dt) for a bird
m aintaining a constant rate of fuel deposition on stopover
will drop as the bird attains progressively larger fuel loads
(f). This relationship is shown for a low -quality (1) and highquality (2) stopover site. Given an expected speed of m ove­
m ent along the m igration route according to the horizontal
broken line, the optimal fuel load for m axim izing speed of
m igration is f-j for departure from the poor site 1 , and f2 for
departure from the high quality site 2. From A le rs ta m
(1991). Reprinted with perm ission from Elsevier Trend Jo u r­
nals.
habitat is more or less widespread, as may be the
case for most passerines. G u d m u n d s s o n et al.
(1991) show how the idea of time minimization must
be applied when only a few, discrete, stopover sites
are at hand, which probably is the case for many
wader and waterfowl species. When a good knowl­
edge exists of the conditions at various potential stop­
over sites (which is rarely the case), detailed
predictions about a complete migratory journey can
be derived.
O P T IM A L R E T U R N JO U R N E Y
The expectation that the speed of migration, or some
other parameter characterizing migration, is opti­
mized, derives from the belief that migration is
shaped by natural selection. Yet, natural selection
maximizes fitness. Thus, it would be nice if it was
possible to evaluate the fitness consequences of
migratory decisions directly. This is exactly what Fig.
9 does by proposing that to each combination of
arrival date and arrival mass a fitness value can
assigned. Building on the knowledge on flight costs
and the physiology of fuel storage discussed above,
the technique of stochastic dynamic programming
(M c N a m a r a & H o u s t o n 1986, M a n g e l & C l a r k
1986, M a n g e l & C l a r k 1988) then allows us to find
the optimal decisions concerning mass gain and
departure for the return journey. Thomas Weber
showed the first results of such an analysis. When
stochasticity was made to increase along the migra­
tion route, the last staging site was sometimes
skipped. Perhaps not surprisingly, departure weights
exceeded the weight necessary to reach the next
stopover site when there was stochasticity in the
expenditure during flight over a given distance. It may
be possible to estimate the extent of this stochasticity
from data on Knots which encounter quite variable
wind conditions between years when migrating from
the Banc d'Arguin to the Wadden Sea in spring
(P ie r s m a & v a n d e S a n t 1992). In unfavourable
years many birds were apparently forced to use an
intermediate stopover site, while arrival in the Wad­
den Sea was delayed.
A
THE GOAL
-^r
Fig. 9. Schem atic presentation of the link between the ultim ate goal (maximizing fitness,
in this case equated to reproductive success on the breeding grounds) and the problem
of how mass gain should be patterned during spring m igration (based on P ie r s m a & E ns
1992).
O P T IM A L D E C IS IO N S A N D S T O C H A S T IC IT Y
Of course, stochastic dynamic programming is not
without its problems either. The technique is non-analytic and often involves many parameters. This makes
it hard to understand the exact causes as well as the
generality of the simulation results. Alasdair Houston
used his work on the trade-off between gaining
energy and avoiding predation to discuss the condi­
tions under which different types of models were most
appropriate to problems of dynamic optimization
(H o u s t o n et at. 1993). When an effect of state (e.g.
fat load) is present and stochasticity is negligible,
analytic solutions can often be found. However, when
the foraging environment is stochastic, simple analyti­
cal models are most likely to fail and stochastic
dynamic programming seems the only theoretical
avenue (e.g. H o u s t o n & M c N a m a r a 1993). The vari­
ous possibilities are depicted in Fig. 10. A striking
example was on the buildup of fat stores prior to
migration in a variable environment. At low reserve
levels risk averse behaviour, meaning preference of a
smaller variance in intake rate at a same mean, is
expected when the latest possible departure date is
still a long way off, but it switches to risk prone behav­
iour, i.e. preference of a higher variance in intake rate
at a same mean, either if reserves are higher or if the
time until departure is reduced (Fig. 11, B e d n e k o f f &
Fig. 10. Diagram depicting the conditions under which differ­
ent types of optim ization m odels are m ost appropriate (Alasdair Houston).
Fig. 11. Graph showing for each com bination o f reserve le v­
els (maximum 80) and tim e steps before final tim e T w hether
a hypothetical m igrant birds should be either risk prone (pre­
ferring a high variance in intake at a sam e mean) or risk
averse (preferring a low variance in intake at a sam e mean).
In the m odel final tim e T represents the last opportunity to
migrate effectively. Successful m igration is only possible
with m aximum reserve levels. Sim ulations are based on
model 2 (prem igratory fattening under predation hazard) of
B e d n e k o f f & H o u s t o n (in press).
80
70
60
= = = - = 3= Risk Prone := 333333= 3=3=33::r
50
Reserves
40
30
£ Risk Prone =
20
Risk Averse
200
400
G00
800
Time
1000
1200
1400
H o u s to n in press). Laboratory experiments by
M o o r e & S im m (1986) demonstrate risk prone behav­
iour in a migrating bird. Indeed, it may well be that
migrating birds offer excellent opportunities for the
study of the risk proneness of foraging behaviour,
since the penalty is not death, but delayed arrival, or
the skipping of a breeding season. That is to say,
migration offers a situation where risk-prone behav­
iour can be advantageous without death being very
likely. L o r ia & M o o r e (1990) and M o o r e (1992)
demonstrated differences in foraging behaviour
between fat and lean migratory birds in field studies
on a stop-over site. The next challenge consists of
actually measuring that the observed foraging behav­
iour is risk prone or risk averse.
Enormous environmental stochasticity was appar­
ent in the study by Martin Morton of the Mountain
White-crowned Sparrow. On the same date the
breeding grounds were sometimes covered by sev­
eral meters of snow, or no snow at all, depending on
the year. For every 2 days delay in breeding, due to
snow conditions at the start of the season, there was
a 1 day delay in departure at the end of the season,
up to a point where departure was not delayed any
further (Fig. 12). There was no link between the date
of egg laying and the onset of moult, but fattening
prior to autumn migration started when moult had
ended.
Fig. 12. Relationship of autum nal m igration departure
schedules of M ountain W hite-crowned Sparrows in relation
to reproductive schedules. Seven years of data (Martin M or­
ton).
M E A S U R IN G T H E F IT N E S S C O N S E Q U E N C E S O F
A R R IV A L D ATE
M odels th a t seek to ca lcu la te th e op tim al tim in g o f the
return jo u rn e y to the b re ed ing g ro u n d s m u st s ta rt
from a re la tio n sh ip be tw ee n fitn e ss and a rriva l date
a n d /o r fitn e ss and arrival co n d itio n (Fig. 9). T hou gh
the lo gic o f the m od els req uire s a d op ting a life history
ap proach, so far, stu d e n ts o f m ig ra tio n ha ve not
attem p te d to a ctu a lly m ea sure th e fitn e ss c o n s e ­
qu en ces o f alte rn a tive m ig ra to ry d e cisio ns. T h is m ay
well be due to the fa c t th a t "tra d itio n a l" life histo ry th e ­
ory ha s co n ce n tra te d on stu d yin g h o w de cisio ns
take n once a ye a r a ffe ct m ortality and rep ro d u ctio n
(see review s in Le s s e l l s 1991 and S t e a r n s 1992).
S ta te -d e p e n d e n t life -h isto ry th e o ry (e.g. M c N a m a r a &
H o u s t o n 1992) see ks to b ridg e the g a p be tw ee n
beha viou ral eco lo g y and life histo ry theory. Still, fit­
ness m ust be m ea sure d. In o rd e r to e stim a te fitne ss
co n se q u e n ce s of m ig ration p a tte rn s w e sh o u ld kno w
the relation be tw ee n re p ro d u ctive succe ss, arrival
date and n u trie n t stores upon arrival. In Fig. 9 the
vie w is ta ke n th a t o p tim al arrival da te is in d e p e n d e n t
o f nu trie n t stores. H ow ever, th e re p ro d u ctive succe ss
as a fu n ctio n o f arrival date is re p re se n te d by th e sum
o f the re p rod uctive va lu e o f th e o ffsprin g and the
residual re p rod uctive va lu e o f th e parents. W hen the
sha pe o f th e fitn e ss cu rve ove r arrival tim e differs
betw een offsprin g and parents, a tra d e -o ff o ve r arrival
tim e exists. W hen, in addition, th e fitn e ss o f th e p a r­
e n t but n o t the o ffsprin g (or vice versa) d e p e n d s on
the n u trie n t sto re s in d e p e n d e n tly o f arrival tim e, than
the op tim al arrival tim e w ill s h ift w ith n u trie n t stores.
F itness co n se q u e n ce s o f arrival date and con dition
ca n n o t be estim a te d re lying on natural variatio n
be tw ee n individuals. J u s t as in th e stu d y o f clutch
size (e.g. T in b e r g e n & D a a n 1990), e xp e rim e n ts
aim ed a t m a n ip u la tin g arrival tim e to e stim a te fitne ss
co n se q u e n ce s fo r both p a re n ts and o ffsp rin g are
needed.
Even though the data are only observational, the
review by D a a n et al. (1989) leaves little doubt that for
single-brooded bird species the reproductive value of
an egg declines in the course of the season. It is
much less clear why this is so. There could be a link
with competitive advantages related to the benefits of
prior residence for either young competing on the
wintering grounds, or adults competing on the breed­
ing grounds, or both. When effects of prior residence
are important, late-fledged chicks are at a clear com­
petitive disadvantage during their first winter. Also, we
would expect that early arriving breeding birds would
be in the best position to take up the best territories
(M y e r s 1981). However, reviewing the evidence,
K e t t e r s o n & N o la n (1983) suggest that "priority of
arrival is not sufficient to establish an indefeasible
claim when the claim is contested by a former owner".
On the one hand there can be no doubt that local sta­
tus is maintained from one year to the next in many
70-
■3>
potential competitors. The experiments of Dan Cristol
(MS) go some way towards this goal. He showed that
experimentally delayed female Red-winged Black­
birds (i.e. held in captivity for a while) were less often
the first to breed in their harem and initiated clutches
later, even though all females were released well
before the first clutches were laid (Fig. 13). These
negative effects on fitness, thought to result from a
decline in social status, were much less severe when
compared to the effects of a similar experiment on
males (in a different population though), explaining
why the polygynous male Red-winged Blackbirds
tend to arrive earlier and much more synchronized
than the females.
b - b:
60
g
501
£
40
<u
w
1J J - -' -o''n
0 BD
Ô
S'
c
30
20 i
10
WHICH MODELS ARE BEST?
0
807060 5040
30 ° C o n tro ls
•
Experim entais: natural arrival
°
Experim entais: delayed arrival
----- R eg re ssio n o f co n tro ls
2010-
0
10
20
30
40
50
60
70
80
Day o f arriva l
Fig. 13. Date of arrival and first egg (in days since first
fem ale arrived) for A) control and B) experim ental fem ales
for 1990 and 1991 combined. Regression equation (dashed
line, y=40.7+0.41x) was com puted for controls only, and is
also shown in B) for com parison. Open circles represent
natural arrival and first egg date for experim ental females.
Solid circles represent the dates of release from captivity
and first egg for the sam e experim ental females. (C r is t o l
MS)
migratory long-lived species like the Oystercatcher.
On the other hand it would seem that preempting
potential usurpers by arriving early on the breeding
grounds is less costly than evicting established
usurpers (ignoring the cost of departing early from the
wintering grounds). The longer Great Tits are experi­
mentally removed from their territory the more likely it
is that they will not succeed in regaining their territory
(K r e b s 1 9 8 2 ).
Thus, methods should be found to measure the
cost of arriving a certain number of days later than
Fig. 9 depicts the timing of arrival as a simple optimi­
zation problem. However, the above discussion
strongly suggests that the concepts of game theory
(M a y n a r d S m it h 1982) need to be applied, since
what matters is not so much the absolute date of
arrival, but the relative timing with respect to the
arrival of potential competitors. This poses further
problems for fitness measurements as well as for
model builders. A t this stage we must ask ourselves
how to proceed as there seems a clear danger that
the enterprise will lead to a forbidding graveyard of
abandoned and underdeveloped models none of
which are rigorously tested. We suggest that the the­
oreticians should see it as their task to make clear
under what conditions the more complex, but suppos­
edly more realistic, dynamic optimization models yield
the same predictions as the more elegant and analyt­
ically tractable rate-maximization models. The empiri­
cists on the other hand should only abandon models
if they consistently fail empirical tests. The model of
A l e r s t a m & L in d s t r ö m (1990) has inspired some
good tests (L in d s t r ö m & A l e r s t a m 1992) and should
inspire more.
THE FULL ANNUAL CYCLE AS AN INDIVIDUAL
DECISION
Noël Holmgren and Anders Hedenström were the
only ones brave enough to tackle the full annual cycle
in their application of dynamic programming to define
the conditions when moult should take place. Pilfering
from the abstract of the paper by Noël and Anders on
this topic (chapter 3 in H o l m g r e n 1993): "The sched­
uling of moult was basically a result of a trade-off
between having a high feather quality during breeding
versus during the non-breeding period. When feather
quality was of major importance for survival, summer
moult tended to be the optimal policy. In the opposite
situation, when feather quality was more important
during breeding, winter moult tended to be optimal. A
high impact of feather quality on survival rate in com­
bination with low costs of moult resulted in biannual
moult (summer and winter moult). Winter moult
became more likely as the survival rate per se
increased. Moult duration, migration costs and repro­
ductive success p e rse were found to have no impact
on the timing of moult." Though the assumption that
feather quality affects fitness seems quite reasona­
ble, the participants of the workshop knew of little
direct evidence, let alone estimates of the magnitude
of the effect. Since then, the work of Dale Clayton and
coworkers on feather lice, heat loss and mate choice
(B o o th et al. 1993) was brought to our attention by
Ellen Ketterson.
POPULATION DYNAMICS OF MIGRATORY
SPECIES
So far we have considered a migration schedule as a
sequence of individual decisions. What we have to
compare our models with though, is the migratory
behaviour of a population of individuals. This sug­
gests that a complete understanding requires that we
are also able to predict the number of birds that follow
a particular migration strategy. This, in turn, forces us
to ask how migratory birds affect each other. We
already touched upon this question when investigat­
ing the effect of arrival time on fitness.
Though there are many ways in which the migra­
tory decisions of one individual may be affected by
the decisions of other individuals (due to the benefits
of flying in a flock for instance; see discussion in
P ie r s m a et al. 1990) the workshop concentrated on
the effects of intraspecific competition on the breed­
ing and wintering grounds. In his book "Populations in
a seasonal environment" F r e t w e l l (1972) graphi­
cally investigated the consequences of densitydependent reproduction on the breeding grounds,
and density-dependent survival on the wintering
grounds. Here, density-dependent means that, with
increasing population density, reproduction in sum­
mer or survival in winter are negatively affected. He
showed that, depending on the shape of the relation­
ships, regulation of the population occurred primarily
in winter, during summer or in both seasons (Fig. 14).
G o s s -C u star d (1980) developed an essentially simi­
lar simulation model. Remarkably, it is only now that
these models are being followed up by attempts to
obtain empirical estimates of the parameters and
more penetrating theoretical investigations, like inclu­
sion of the cost of migration, and attempts to derive
the population model from knowledge of the behav­
iour of individuals.
John Goss-Custard presented the results of a col­
laborative effort to amass all available data on breed­
ing and survival of the European populations of the
Oystercatcher (G o s s -C u star d et al. in press (a);
G o s s -C u star d et at. in press (b)). Assuming strong
Fig. 14. Graph of a very simple model of population regula­
tion in a seasonal environment. Production occurs during
the breeding season with rate b, while deaths are confined
to winter and occur with rate d. The population is stable if
d=b/(1+b), i.e. if the death rate equals the “replacement
rate” g. The equilibrium autumn population size N(1 +b)eq is
found by plotting both g and d against the autumn population
size N(1+b). From F r e t w e l l (1972).
density dependence in summer, it was calculated that
populations would decline following habitat loss on
the wintering grounds, if wintering densities exceeded
those where winter mortality became density-depend­
ent. With steeper density dependence more birds
would be lost. Though the database is impressive, the
most important parameters, i.e. those describing
important aspects of the density-dependent relation­
ships, are least well known. To derive these popula­
tion parameters from knowledge of the behaviour of
individuals (see G o s s -C u s tar d & D u r e l l 1990) a
game theory model of how Oystercatchers distribute
themselves over the musselbeds in the Exe estuary
was constructed. From this, a relationship between
winter mortality and population size could be derived.
The conceptual framework of the model presented by
Paul Dolman was essentially similar, but its imple­
mentation differed. Instead of splitting the problem
into several submodels, each with empirically esti­
mated parameters, Paul Dolman presented a general
simulation model incorporating all the processes
believed to be relevant (Fig. 1 5, S u t h e r l a n d & D o l ­
m an in press). Below, the pros and cons of these con­
ceptually similar models are investigated through a
close examination of the components, confronting the
assumptions with the relevant insights of other partic­
ipants to the workshop.
DENSITY-DEPENDENT REPRODUCTION
Density-dependent reproduction will occur when with
increasing population size an increasing proportion of
birds settle in habitat of poor quality or do not breed,
FRAMEWORK
B R E E D IN G
GROUNDS
WINTERING
GROUNDS
Fig. 15. Diagram depicting the presumed nature of the den­
sity-dependent processes on the breeding and wintering
grounds (Paul Dolman).
the so-called buffer effect ( K lu ijv e r & T in b e rg e n
1953). Detailed long-term studies of the American
Redstart and the Blackthroated Blue Warbler by Rich­
ard Holmes and colleagues provided evidence for
variation in habitat quality affecting fitness, leading to
competition for high quality habitats, where food is
abundant and nest predation low (H o lm e s e t al.
1992; S h e r r y & H o lm e s 1992). In Oystercatchers,
intense competition exists for territories of high qual­
ity, where transport costs of food to the chicks are low
(Ens e t al. 1992). E n s e t al. (M S2) suggest that
potential recruits face a trade-off between immedi­
ately settling in a territory of poor quality or queuing
for a territory of high quality. Thus, the only freedom
left to the newly arriving nonbreeders when faced with
intense competition is the decision on how to cut their
losses: losing fitness in time (by queuing for a high
quality territory) or losing fitness in space (by settling
in a poor quality territory). Queuing must result from
the advantage enjoyed by prior residents in the com­
petition for space and the impossibility of maintaining
high status over a large area (E ns e t al. MS1).
Despite the evidence for competition (see also the
section on the fitness consequences of arrival date),
the shape of the curve relating reproduction to the
number of potential settlers remains elusive.
Whereas differences between individuals are thought
to be of critical importance in models describing density-dependent winter survival, such individual differ­
ences were not explicitly incorporated in the
submodels of John Goss-Custard and Paul Dolman
dealing with density-dependent reproduction. There
is a cle ar need to de rive th e d e n sity -d e p e n d e n t re la ­
tio n s h ip s from the a c cu m u la tin g kn o w le d g e on the
m ech an ism o f social c o m p e titio n du rin g th e bre ed ing
season.
DENSITY-DEPENDENT W IN TER SURVIVAL
W in te r su rviva l will be d e n sity -d e p e n d e n t w h en food
stocks can be de pleted , or w h en in te rfe re n ce forces
an in cre a sin g n u m b e r o f b ird s in to h a b ita ts o f poor
qu ality w ith in cre a sin g p o p u la tio n size. E vid e n ce for
m assive d e p le tio n o f foo d stocks by va rio u s spe cie s
of w a te rfo w l on m ig ration w a s p ro vid ed by M en no bart
van E erden. In som e e xa m p le s fo o d w a s d e pleted to
a sam e th re sh o ld level ea ch year, b e fore th e birds
m ove d on, w h ile in o th e r e xa m p le s th e th re sh o ld level
w as va ria b le (van E e rd e n 1984; D r e n t & P rin s
1987). T he la tte r m ay have been d u e to va ria b ility in
th e fo o d su p p ly a t p re ce d in g sta g in g sites, a n d /o r the
need to a rrive on th e w in te rin g s ite a t a ce rta in date.
In contrast, w in te rin g O yste rca tch e rs rare ly rem ove
m ore than 30% o f the stan din g sto c k o ve r the w inter,
b ut in te rfe re n ce is a w e ll-e sta b lish e d p h en om en on ,
in clud in g the o b se rva tio n th a t in d ivid u a ls diffe ring in
d o m in a n ce s u ffe r to a d iffe re n t d e g re e from the p re s ­
en ce o f co n sp e cifics (Ens & G o s s - C u s ta r d 1984;
G o s s - C u s ta r d e t al. 1984). S u b s e q u e n t th e o rie s on
th e d istribu tion o f un eq ua l c o m p e tito rs have coined
th e term "co m p e titive a b ility" to d e scrib e th e se d iffe r­
en ce s betw een in d ivid u a ls ( S u t h e r la n d & P a r k e r
1985; P a r k e r & S u t h e r la n d 1986). H ow ever, this
c o n ce p t m ay n o t a p p ly in th e fie ld (see a p p e n d ix I).
For instance, it is hard to id e n tify th e b e tte r c o m p e tito r
w ith re sp e ct to the a cq u isitio n o f foo d w h en th e re is a
tra d e -o ff be tw ee n fo ra g in g a b ility a n d fig h tin g ability.
G ood fo ra g e rs m a y have h ig h e r su rviva l cha nce s
w hen food s u p p lie s are se ve re ly de pleted , w h ile good
fig h te rs can e xclu d e o th e r birds from th e b e st fe e d in g
areas. Potentially, such a tra d e -o ff m ig h t exp la in w hy
S h e r r y & H o lm e s (MS) did n o t find s ig n ific a n t s u rv i­
vorsh ip diffe re n ce s be tw ee n A m e rica n R e d sta rts w in ­
te rin g in d iffe re n t ha bita ts, eve n th o u g h rem oval
e xp e rim e n ts
sh o w e d
th a t
m ales
d e sp o tica lly
e xclud ed fe m a le s from certain h a b ita ts ( M a r r a e t al.
1993). F urthe rm o re, social d o m in a n c e is not n e c e s ­
sa rily so le ly de te rm in e d by re la tive fig h tin g ability, but
m a y a lso be de te rm in e d by p rio r re side nce, e s p e ­
cia lly w h en in d ivid u a ls sh o w a stro n g site fidelity.
P rese nt e vid e n ce ce rta in ly a llo w s th e su g g e stio n tha t
O yste rca tch e rs w in te rin g on m u sse lb e d s "q ue ue" fo r
p o sitio n s o f high local d o m in a n ce (E ns & C a y fo r d , in
press). T h e e xp e rim e n ts o f Dan C ristol on D a rk-eyed
Ju n co s le ft little do ub t th a t be h a vio u ra l d o m in a n ce is
in flue nced by situ a tio n a l factors, like a rrival date
(prior re sid e n ce in crea ses do m in a n ce , C r i s t o l e t al.
1990), h u n g e r (lean birds a re in itia lly d o m in ant, C r is ­
t o l 1992) and th e fa m ilia rity w ith a m ore d o m in a n t
bird, as w ell as a ttrib u te s o f th e b ird s like sex.
D etailed stu d ie s o f th e W h ite -th ro a te d S pa rrow s
depletion. T h is m ad e it n e ce ssa ry to assu m e th a t the
ideal fre e distrib u tio n w as reached a t ea ch tim e step.
T h e co n d itio n s un d e r w h ich th is a s su m p tio n holds
have been th e o re tica lly in ve stig a te d by
et
al. (1988, 1991). E m pirical e vid e n ce s u g g e stin g clo se
tra ckin g o f flu ctu a tio n s in h a b ita t q u a lity o ve r a fa irly
w ide g e o g raph ica l a re a w as pro vid ed by M e n n o b a rt
van E erd en throu gh an im p re ssive e xa m p le on Teal.
In som e yea rs one fifth o f the E uro pe an p o pu latio n
co n ce n tra te d in the n a ture re se rve th e O ostvaa rde rspla ssen , w h ile in o th e r ye a rs n u m b e rs w ere
ve ry low, in line w ith flu ctu a tio n s o f the foo d sup ply
re lated to the m a n a g e m e n t o f th e w a te r le ve ls (Fig.
17).
Bernstein
X1000
Total population attem pting to settle
Fig. 16. The proportion of O ystercatchers predicted to die on
the Exe estuary in w inter in relation to the total num ber arriv­
ing there in autum n (means of several m odel runs, bars rep ­
resent one SE). The threshold intake rate values required
for a bird to rem ain on the estuary were either 200 mg
AFDM /5 min (open circles) or 150 mg AFDM /5 min (solid cir­
cles). The solid triangles show the effects of rem oving the
best 30% o f the total m ussel area, assum ing a threshold
intake rate of 150 mg AFDM /5 min. From G o s s - C u s ta r d et
al. (in press (e)).
underline the importance of site-dependency as well
as age and sex in determining dominance in the field
(Piper & Wiley 1989; Piper & Wiley 1990; Piper
1990).
The presented population models did not incorpo­
rate these details of the competitive process. John
Goss-Custard derived the relationship between over­
winter mortality and autumn population size from a
game theory model of how Oystercatchers differing in
their susceptibility to interference and in foraging effi­
ciency distribute themselves over the musselbeds of
the Exe (Goss-Custard e t at. in press (c,d & e)). The
model assumed that each individual would move to
the musselbed where it would achieve the highest
intake rate and that there were no costs to moving, in
line with the essence of the assumptions of the ideal
free distribution model (Fretwell & Lucas 1970)
according to Kacelnik e t at. (1992). Though the
model will not be the last word on the subject (GossCustard ef a/.in press (e)) the exercise demonstrated
beyond doubt that the precise shape of the densitydependent winter mortality can be derived from
knowledge of the behaviour of individuals in combina­
tion with measurements of the distribution of
resources (Fig. 16). In the model presented by Paul
Dolman, density dependence in winter mortality not
only resulted from interference, but also from prey
Fig. 17. The annual peak num ber of Teal in the freshw ater
marsh O ostvaardersplassen, Netherlands, plotted against
the annual seed production of Ranunculus sceleratus. Large
differences between years are due to m anipulations of the
w ater level. From v a n E e rd e n & M u n s te r m a n (M S).
T H E C O S T O F M IG R A T IO N
M ig ration m u st be costly, but w h a t a re th e co sts and
ho w high are th e y? F or p o pu latio n m o d e ls o f m ig ra ­
to ry birds it is im p o rta n t to kno w if m o rta lity on m ig ra ­
tion sho uld be m od elled a s in cre a sin g o n ly w ith
dista nce , o r as being de n sity d e p e n d e n t as well. It
has on ly re ce n tly be com e po ssib le to e stim a te the
co n trib u tio n o f the e n e rg y s p e n t on m ig ra tio n to the
an nu al e n e rg y b u d g e t ( D r e n t & P ie r s m a 1990;
P ie r s m a e t al. 1991; W ie r s m a & P ie rs m a , 1994).
W ie r s m a & P ie r s m a (1994) have used h e ated ta xide rm ic m ou nts to p re d ict sta n d a rd o p e ra tive te m p e ra ­
ture, or, in verse ly, m a in te n a n ce m etab olism , for
d iffe re n t m icro h a b ita ts and w e a th e r con ditions. T his
m e th o d o lo g y allow ed them to co n clu d e th a t if island ica Knots, the s u b sp e cie s th a t ha b itu a lly w in te rs on
the in te rtida l m ud flats o f W e ste rn E urope, w e re to
m ove to W e s t A frica, h a b itu a lly occu p ie d b y ca n u tu s
K nots, th e y w o u ld in cu r a saving o f 1.1 W on th e ir
m a in te n a n ce m etab olism and pa y an extra 0 .1 -0 .2 W
to c o ve r th e co st o f tra ve l (Fig. 18). T h e p ro blem is
Fig. 18. M aintenance m etabolism on the tem perate Dutch W adden Sea and on the tropical
Banc d'Arguin, Mauritania, W est Africa predicted for islandica Knots on the basis of m eteoro­
logical data and m icrohabitat specific equations. These equations were calibrated with the
use of heated taxiderm ie m ounts. From W ie rs m a & P ie rs m a (1994).
that it is not clear how energy expenditure relates to
survival and prospects for reproduction.
To calculate how birds from different breeding
grounds should distribute themselves over wintering
grounds at different distances, Paul Dolman simply
assumed the existence of a critical intake rate, neces­
sary to achieve the breeding grounds. The critical
intake rate was thought to increase linearly with dis­
tance between the wintering and breeding grounds.
Instead of assuming that birds with low intake rates
do not reach the breeding grounds at all, it seems
more likely that they will reach the breeding grounds
later, probably at the cost of a reduction in reproduc­
tive success (Fig. 9). Both assumptions lead to density-dependence in either mortality or reproduction
following migration for which there exists no empirical
evidence whatsoever. An alternative assumption
could therefore be that the crossing of a certain dis­
tance carries a fixed mortality risk. Estimating this risk
is difficult. Peter Evans concluded that for several
species of waders the actual mortality during the
migratory journey is low relative to the mortality on
the wintering and the breeding grounds (Pienkowski
& Evans 1985; Evans 1991). The reverse seems to
be the case for the small migratory passerines stud­
ied by Dick Holmes and colleagues (Holmes et al.
1989; Holmes & Sherry 1992). Martin Morton
reported that there existed no differences in age-specific survivorship curves between a migrant and a
nonmigrant subspecies of the White-Crowned Spar­
row. A problem with all these survival estimates is
that it is not possible to separate lack of fidelity to the
wintering or breeding area from mortality. The only
conclusive solution to this problem is to study a signif­
icant part of the population throughout the year, as
can be done in some species of geese (e.g. Owen &
Black 1989; Ebbinge et al. 1991). This will also help
in testing the suggestion of Pienkowski & Evans
(1985) that an important reason for minimizing the
flight distances on the first migration is to reduce the
risk of failing to complete the journey to a suitable
“wintering" site, i.e. adults may not suffer high losses
on migration, but the inexperienced juveniles may.
PREDICTING THE EFFECTS OF HABITAT
CHANGE
An im p o rta n t in cen tive in seve ral o f the p re vio usly
d iscusse d stu d ie s w as th e d e sire to u n d e rsta n d and,
if possible, p re d ict the e ffe c t o f m a n -in d u ce d ha bita t
cha ng es on the p o pu latio n dyn a m ics o f the m ig ratory
birds. S uch m a n -in d u ce d ch a n g e s ra n g e from local
reclam atio n o f so m e o f the m u d fla ts in an e stu a ry to
global clim a te change. A u n ifyin g th e m e in th e stud ies
o f D ick H o lm es and c o -w o rke rs fo r in sta n ce has been
th e de sire to id e n tify th e ca u se s o f the dra m atic
d e clin e in the n u m be rs of n e o tro p ica l m ig ra n t p a s s e r­
ines
in press).
et
al. in pre ss (a) used th e ir p o p u la tio n m od el to c a lc u ­
late the e ffe ct o f loss o f w in te rin g h a b ita t on th e e q u i­
librium p o pu latio n size o f th e E u ro p e a n p o pu latio n of
O yste rca tch e rs
et al. in pre ss (b).
T hey a rrived a t the startlin g co n clu sio n th a t the
coa stal su b p o p u la tio n s w o u ld be s tro n g ly ne g a tive ly
affected by h a b ita t loss, w h ile n e g a tive e ffe cts on the
inlan d b re ed ing s u b p o p u la tio n s o n ly b e ca m e a p p a r­
e n t a fte r th e coa stal su b p o p u la tio n s w e re de cim a te d.
T h e re is a cle a r need to te s t th e u n d e rlyin g a s s u m p ­
tio n th a t coa stal and inland birds co m p e te on equal
te rm s on the w interin g grounds.
(Holmes & Sherry 1992; Sherry & Holmes
1992; Sherry & Holmes
Goss-Custard
(Goss-Custard
When the ESS migration model of Sutherland &
Dolman (in press) was used to predict the effect of
habitat change on the numbers of birds using differ­
ent migration routes, it was assumed that the birds
can adapt to the new circumstances, culturally or
genetically, i.e. a new ESS can be achieved.
Berthold & Querner (1981) and Berthold (1988)
have shown the existence of genetic variation in the
migratory behaviour of the Blackcap. Berthold et al.
(1992) provide evidence that this species has evolved
a new migration route over the past 30 years. Instead
of assuming that a new ESS is reached in response
to habitat loss without problems, Paul Dolman
(unpublished) therefore theoretically investigated the
conditions under which a species with a genetically
determined migration strategy could survive a drastic
change of habitat. The model showed that for the
migration strategy to track the changing environment,
it was important to have sufficient genetic variance in
the population. The model also showed that assortative mating can allow a much more rapid response,
allowing tracking even when the level of genetic vari­
ation is initially very low.
CONCLUSIONS
S O M E IM P O R T A N T A N D H O P E F U L L Y A M E N A B L E
PR O BLEM S
W hile d iscussin g the va rio u s m ig ra to ry d e cisio n s in
turn, a host o f p ro blem s w a s e n cou ntere d. It see m s
pru de nt to m ake a sh o rt list o f the m ost im p ortan t
ones.
• T hou gh e n e rg y is rig htly d e scrib e d as the fire o f life
1961), an anim al m u st also retain w a te r
and protein ba la nce during th e m ig ra to ry flight, to p ­
ics th a t d e se rve and a ttra ct in crea sing attention.
• T here can be little do u b t th a t the po w e r cu rve is
cen tra l to pre d ictio n s on speed, a ltitu de and range
o f th e m ig ratory flight, its e lf a crucial e le m e n t o f any
m ig ration sche du le. T hou gh th e re w a s general
a g re e m e n t th a t the po w e r curve m u st be U -shaped,
it w as fe lt th a t m uch m ore em p irica l w o rk w as
needed to confirm th e pre cise sh a p e fo r a given
species, a ne ce ssa ry p re re q u isite fo r reliab le q u a n ­
tita tive predictions. O n th e basis o f e m p irica l e v i­
d e n ce it is im p ossib le to d e cid e be tw ee n the
m od els o f
(1989) o r
(1988),
b u t th e fo rm e r has fe w e r and m ore e a sily m ea sure d
pa ra m e te rs tha n the latter.
• M ig ration m u st be costly, but w h a t are th e costs
and how high are the y? It is in tu itive ly a p pe aling to
assu m e th a t high e n e rg y e x p e n d itu re re d u ce s s u r­
vival (e.g.
e t al. 1990), but nothing con clu sive
is known. For po pu latio n m od els o f m ig ra to ry birds
it is im p o rta n t to kn o w if m o rta lity on m igration
s h o u ld be m o d elled as fixed fo r a given distance, or
as d e n sity depe nd ent.
(Kleiber
Pennycuick
DAAN
Rayner
• T h o u g h the re is an in cre a sin g a w a re n e ss th a t p o p ­
ulatio n m od els sh o u ld be de rive d from th e in te ra c ­
tio n s be tw ee n individuals, th e m o d e ls discu sse d
con tain ed a ssu m p tio n s ab o u t th e c o m p e titive p ro c ­
ess th a t m a y be un tenable. O ne of th e a u th o rs o f
th is re p o rt fe e ls ra th e r s tro n g ly th a t e m p irica l s tu d ­
ies o f the c o m p e titive p ro ce ss w o uld b e n e fit from a
c la rifica tio n o f c o n ce p ts (see ap p e n d ix I).
M IG R A T O R Y B IR D S P R O V ID E G O O D R E S E A R C H
SYSTEM S
In th e a b ove w e have listed w h a t w e re co n sid e re d to
be th e m o st im p o rta n t problem s, w ritten from the p e r­
sp e ctive o f a scie n tist w h o has decided, fo r w h a te ve r
reason, th a t solving th e m yste ry o f bird m ig ra tio n is
w h a t he w a n ts to do w ith his life. H ow ever, scie ntists
w a ntin g to solve a d iffe re n t problem sho uld in no w ay
fee l exclud ed. If the stu d y o f m ig ration s tra te g ie s is to
becom e a se rio u s u n de rta king d ire ct fitn e ss m e a s u re ­
m ents m ust be take n. T hus, the re is a cle a r need fo r
a clo se r link betw een m ig ration and bre ed ing b io lo g i­
cal stud ies o f the sam e po pu latio ns. In fact, it is even
p o ssib le th a t scie ntists w ith d iffe re n t in te re sts w ill find
th a t m ig ra to ry birds are p a rticu la rly su ita b le fo r a n s ­
w e ring th e ir questions. T h e fo llo w in g su g g e stio n s
cam e forw ard:
• B irds p re pa ring fo r m ig ration m ay o ffe r go od o p p o r­
tu n itie s fo r stud ying w h e th e r forag in g b e h a vio u r is
risk prone or risk averse. T h e pro blem see m s th a t
field w o rke rs lack a sea rch im a ge fo r such b e h a v ­
iour: the th e o re ticia n s sho uld co m e forw a rd w ith
su g g e stio n s o f w h a t to look for.
• M ig ratory birds m ay o ffe r be tte r o p p o rtu n itie s fo r
stud ying sea son al tim in g o f m o u lt and rep ro d u ctio n
than se d e n ta ry birds. A tte n tio n should fo cu s on
arrival tim e and the e xp e rim e n ta l m a n ip u la tio n o f
th a t arrival tim e.
PRO SPECTS
T h e w o rksh o p le ft the p a rticip a n ts w ith little d o u b t th a t
the stu d y o f bird m igration, in p a rticu la r th e stu d y o f
m ig ration sche du les, is rap id ly g a in ing m om e ntum
th ro u g h th e a p p lica tio n o f th e sa m e id ea s and te c h ­
niques, first d e velope d by eco no m ists, so s u c c e s s ­
fu lly a p plied in o th e r fie ld s o f b e ha viou ral eco lo gy. To
m aintain m om e ntum it is n e ce ssa ry th a t th e m ore
e m p irica lly m in de d and the m ore th e o re tic a lly m inded
stud ents o f m ig ration c o n tin u e to rem a in in clo se c o n ­
tact. T o m in im ize th e d a m a g e o f exce ssive social
co m p e titio n it is n e ce ssa ry th a t this co n ta ct is n o t only
e sta blished throu gh the scie n tific literature, but ab ove
all th ro u g h p e rson al con tacts. A clo se c o o p e ra tio n
be tw ee n th e o rists and em p iricists w o u ld fit the tra d i­
tion o f b e h a vio u ra l eco lo gy. It w ill be im p o rta n t th o u g h
to avo id re p ea ting a com m o n m ista ke o f m an y b e h a v ­
ioural e co lo gists, w h o have ne g le cte d th e stu d y o f
m ech an ism a t th e e xp e n se o f th e stu d y o f th e e v o lu ­
tio n a ry co sts and b e ne fits
1993).
(Huntingford
REFERENCES
A le r s ta m , T., 1979. Optim al use of wind by m igrating birds:
com bined drift and overcom pensation. J. theor. Biol.
79: 341-353.
— , 1991. Bird flig h t and optim al m igration. Trends Ecol.
Evol. 6: 210-215.
A l e r s t a m , T., G .A . G u d m u n o s s o n & B. L a r s s o n , 1993.
Flight tracks and speeds of Antarctic and Atlantic sea­
birds: radar and optical m easurem ents. Phil. Trans. R.
Soc Lond. B 340: 55-67.
A le r s ta m , T. & Â L in d s tr ö m , 1990. Optimal bird migration:
the relative im portance of tim e, energy, and safety. In:
E. G w in n e r. Bird Migration, Physiology and Ecophysiology. Berlin: Springer Verlag: 331 -351.
B a k e r , R.R., 1978. The Evolutionary Ecology of Animal
Migration. Holm es and Meier, New York.
BEDNEKOFF, P.a. & A.I. H o u s t o n , in press. Risk-sensitive
foraging, m ass-dependent predation, and prem igratory
fattening. Ecology.
B e r n s te in , C., A.J. K a c e ln ik & J R. K re b s , 1988. Individual
decisions and the distribution o f predators in a patchy
environm ent. J. Anim. Ecol. 57: 1007-1026.
B e r n s te in , C ., A.J. K a c e ln ik & J.R. K re b s , 1991. Individual
decisions of predators in a patchy environm ent. II. The
influence of travel costs and structure of the environ­
m ent. J. Anim . Ecol. 60: 205-225.
B e r t h o l d , P., 1988. Evolutionary aspects of m igratory
behavior in European warblers. J. evol. Biol. 1: 195209.
B e r t h o l d , P., A.J. H e lb ig , G. M o h r & U. Q u e r n e r , 1992.
Rapid m icroevolution of m igratory behaviour in a wild
bird species. Nature 360: 668-670.
B e r tv io ld , p . & U. Q u e r n e r , 19 8 1 . G enetic basis of m igra­
tory behavior in European warblers. Science 212: 7779.
B ie b a ch , H. W. F r ie d r ic h , G. H eine , L. J e n n i, S. J e n n iE ie rm a n n & D. S c h m id l, 1991. The daily pattern of
autum n bird m igration in the northern Sahara. Ibis 133:
414-422.
B ie s e l, W. & W. N a c h t ig a ll, 1987. Pigeon flight in a wind
tunnel. IV. therm o-regulation and water homeostasis.
J. comp. Physiol. B 157: 117-128.
B o o t» , D.T., D.H. C la y t o n & B .A . F lo c k , 1993. Experim en­
tal dem onstration o f the energetic cost o f parasitism in
free-ranging hosts. Proc. R. Soc Lond. B 25 3: 125-129.
B r u d e r e r , B., in press. Nocturnal bird migration in the
Negev - A tracking radar study. Ostrich 65.
B r u d e r e r , B., L.G. U n d e r h i l l & F. U e c h ti, in press. A lti­
tude choice of night m igrants in a desert area predicted
by m eteorological factors. Ibis 136.
B r u d e r e r , B. & E. W e itn a u e r , 1972. Radarbeobachtungen
über Zug und N achtflüge des M auerseglers (Apus
apus). Rev. Suisse Zool. 79: 1190-1200.
C a rm i, N., B. P in s h o w , W . P. P o r t e r & J. J a e g e r, 1992.
W ater and energy lim itations on flight duration in small
m igrating birds. Auk. 109: 268-276.
C h r is t y , J.H. & M. S a l m o n , 1984. Ecology and evolution of
m ating system s of fiddler crabs (genus Uca). Biol. Rev.
59: 483-509.
C r is t o l , D.A., 1992. Food deprivation influences dom i­
nance status in dark-eyed juncos, Junco hyemalis.
Anim . Behav. 43: 11 7 -1 2 4 .
C r is t o l , D.A., MS. Experim entally delayed arrival of early
fem ale red-winged blackbirds.
C r is t o l , D.A., V .N . j r N o l a n & E.D. K e t t e r s o n , 1990.
Effect of prior residence on dom inance status o f dark­
eyed juncos, Junco hyemalis. Anim . Behav. 40: 5 8 0 586.
D a a n .S., C. D ij k s t r a , R.H. D r e n t & T. M e ij e r , 1989. Food
supply and the annual tim ing of avian reproduction. In:
Proceedings of the XlX th International Ornithological
Congress, pp. 3 9 2 -4 0 7 . O ttawa, O ttaw a Press I.
D a a n , S., C. D ij k s t r a & J.M . T in b e r g e n , 19 90. Family plan­
ning in the kestrel (Falco tinnunculus): the ultim ate con­
trol of covariation of laying date and clutch size
Behaviour 114: 8 3 -1 1 6 .
D r e n t , R.H. & T. P ie r s m a , 1990. An exploration of energet­
ics of leap-frog m igration in arctic breeding waders, pp.
3 9 9 -4 1 2 In: E. G w in n e r . Bird m igration: physiology and
ecophysiology. Springer-verlag, Berlin.
D r e n t , R.H. & H.H.T. P r in s , 1987. The herbivore as a p ris ­
o n e r o f its fo o d s u p p ly . C h a p te r 9 (pp. 1 3 1 -1 4 7 ). In: J.
A n d e l et al. D is tu rb a n c e in G ra s s la n d s . Dr. W .
J u n k P u b lis h e rs , D o rd re ch t.
E b b in g e , B .S ., J .B . v a n B ie z e n & H. v a n d e r V o e t , 1991.
E s tim a tio n o f a n n u a l a d u lt s u rv iv a l ra te s o f B a rn a cle
G e e se Branta leucopsis u sin g m u ltip le re s ig h tin g s o f
m a rke d in d ivid u a ls. A rd e a 79: 7 3 -1 1 2 .
E e r d e n , M.R. v a n , 1984. W aterfowl m ovem ents in relation
to food stocks. In: P R. E v a n s , J.D. G o s s -C u s t a r d &
W .G . H a l e . Coastal waders and w ildfow l in winter, pp.
8 4 -1 0 0 . Cam bridge Univ. Press, Cam bridge.
E e r d e n , M.R. v a n & M.J. M u n s t e r m a n , MS. Seed density
van
and waterlevel fluctuations determ ining food uptake
and bird num bers in Teal (Anas crecca) stopping over
at two freshw ater m arshes in the Netherlands.
E n s , B.J. & J. C a y f o r d , in press. Feeding with other oystercatchers. In: J.D. G o s s -C u s t a r d . The behavioural
ecology o f the oystercatcher. Oxford Univ. Press,
Oxford.
E n s , B.J. & J.D. G o s s -C u s t a r d , 1984. Interference among
oystercatchers, Haem atopus ostralegus, feeding on
m ussels, M ytilus edulis, on the Exe estuary. J. Anim.
Ecol. 53: 2 1 7 -2 3 1 .
E n s , B.J., H. vd Jeugd & D. Heg. M S 1 . Territory acquisition
in the oystercatcher: vital preparations for a breeding
career.
E n s , B .J., M, K e r s t e n , A . B r e n n in k m e ij e r & J.B . H u l s C h e r , 1992. T e rrito ry q u a lity , p a re n ta l e ffo rt a n d re p ro ­
d u c tiv e s u c c e s s o f o y s te rc a tc h e rs
( Haem atopus
ostralegus). J. A n im . E col. 6 1 : 7 0 3 -7 1 5 .
E n s B.J., T. P ie r s m a , W.J. W o l f f & L. Z w a r t s , 1990.
Hom eward bound: problem s waders face when m igrat­
ing from the Banc d'Arguin, M auritania, to their north­
ern breeding grounds in spring. Ardea 78: 1-16.
E n s , B.J., F.J. W e is s in g & R.H. D r e n t , MS2. The despotic
distribution and deferred m aturity: two sides of the
sam e coin.
E v a n s , P .R ., 1991. Seasonal and annual patterns of m ortal­
ity in m igratory shorebirds: som e conservation im plica­
tions. In: C.M. P e r r in s , J..D L e b r e t o n , G .J.M H ir o n s .
Bird Population Studies: relevance to conservation and
m anagem ent. Oxford: Oxford University Press: 3 4 6 359.
F r e t w e l l , S., 19 72. Populations in a seasonal environm ent.
Princeton: Princeton University Press.
— , 1980. Evolution o f m igration in relation to factors regulat­
ing bird numbers. In: A. K e a s t , E.S. M o r t o n . Migrant
Birds in the Neotropics: Ecology, Behavior, D istribu­
tion, and Conservation. W ashington: Smithsonian
Press: 517-527.
F r e t w e ll, S. & H.L. Lucas j r , 1970. On territorial behaviour
and other factors influencing habitat distribution in
birds. Acta, biotheoretica 19: 16-36.
G a u t h r e a u x , S.A., 1982. The ecology and evolution of
avian migration systems. In: D.S. F a r n e r , J.R. K in g ,
C.R. PARKES. Avian Biology, Vol IV. New York: A ca ­
demic Press: 93-168.
GOSS-CUSTARD, J.D., 1980. Com petition for food and inter­
ference am ong waders. Ardea 68: 31-52.
G o s s -C u s t a r d , J.D . & S . E.A. D u r e l l , 1988. The effect of
dom inance and feeding m ethod on the intake rates of
O ystercatchers, Haem atopus ostralegus feeding on
m ussels, M ytilus edulis. J. Anim. Ecol. 57: 827-844.
G o s s -C u s t a r d J.D . & S.E.A. D u r e l l , 1990. Bird behaviour
and environm ental planning: approaches in the study
of w ader populations. Ibis 132: 273-289.
G o s s -C u s t a r d , J.D ., R.T. C l a r k e & S . E.A. D u r e l l , 1984.
Rates of food intake and aggression o f oystercatchers
H aem atopus ostralegus on the m ost and least pre­
ferred m ussel M ytilus edulis beds of the Exe estuary. J.
Anim. Ecol. 53: 233-245.
G o s s -C u s t a r d , J.D., R.G. C a l d o w , R.T. C la r k e & A.D.
WEST, in press (e). Deriving population param eters
from individual variations in foraging behaviour: II.
Model tests and population param eters. J. Anim. Ecol.
G o s s -C u s t a r d , J.D., R.G. C a l d o w , R.T. C l a r k e , S . E.A.
D u r e l l & W .J. S u t h e r l a n d , in press (c). Deriving pop­
ulation param eters from individual variations in forag­
ing behaviour: I. Empirical gam e theory distribution
m odel of oystercatchers H aem atopus ostralegus fe e d ­
ing on m ussels M ytilus edulis. J. Anim. Ecol.
G o s s -C u s t a r d , J.D ., R.T. C l a r k e , K.B. B r ig g s , B.J. E n s ,
K-M. Exo, C .J. S m it , A.J. B e in t e m a , R.W.G. C a l d o w ,
C a t t DC , N. C l a r k , S . E.A. D u r e l l , M.P. H a r r is , J.B.
H u l s c h e r , P .L. M e in in g e r , N. P ic o z z i , R. P r y s -J o n e s
& U. S a f r ie l , in press (a). Modelling the effect of habi­
tat loss on the European population of the oystercatcher Haem atopus ostralegus: I. Estim ating m odel
param eters. J. Appl. Ecol.
G o s s -C u s ta rd , J.D., R.T. C la rk e , S.E.A. D u r e ll, R.W.G.
C a ld o w RW G & B.J. Ens, in press (b). Predicting the
effect of winter habitat loss on the European population
of the oystercatcher Haem atopus ostralegus'. II. Model
predictions. J. Appl. Ecol.
G o s s -C u s t a r d J.D., A.D. W e s t , R .W .G . C a l d o w S.E.A.
D u r e l l & S. M c G r o r t y , in press (d). An empirai opti­
mality m odel to predict the intake rates of oystercatch­
ers (Haem atopus ostralegus) feeding on mussels
(Mytilus edulis). Ardea
G r a f e n , A., 1987. The logic of divisively asym m etric con­
tests: respect for ow nership and the desperado effect.
Anim. Behav. 35: 4 6 2 -4 6 7 .
G u d m u n d s s o n , G..A, Â. L in d s t r ö m & T. A l e r s t a m , 1991.
Optimal fat loads and long distance flights by migrating
Knots Calidris canutus, Sanderlings C. alba, and T urn­
stones A renaria interpres. Ibis 133: 140-152.
G w in n e r , E., 1990. Circannual rhythm s in bird m igration:
control of tem poral patterns and interactions with pho­
toperiod. In: E. G w in n e r . Bird Migration, Physiology
and Ecophysiology. Berlin: Springer Verlag: 257-268.
H e d e n s t r ö M, A. & T. A l e r s t a m , 1992. Climbing perform ­
ance of migrating birds as a basis for estimating limits
for fuel-carrying capacity and muscle work. J. exp. Biol.
H e d e n s trö M , A. & T. A le r s ta m , 1994. Optimal clim bing
flight in m igrating birds: predictions and observations of
knot and turnstone. Anim. Behav., in press.
H o lm e s , R.T. & T.W . S h e r r y , 1992. Site fidelity of m igratory
w arblers in tem perate breeding and neotropical w inter­
ing areas: im plications for population dynam ics, habitat
selection and conservation: 5 6 3 -5 7 5 . In: J.M. H a g a n III
& D.W. J o h n s to n . Ecology and C onservation of Neo­
tropical M igrant Landbirds. Sm ithsonian Press, Wash.
D.C.
H o lm e s , R .T., T .W . S h e r r y & L. R e its m a , 19 89. Population
structure, territoriality, and overw inter survival of two
m igrant warbler species in Jamaica. Condor 91: 5 4 5 561.
H o lm e s , R.T., T.W. S h e r r y , P.P. M a r r a & K.E. P e t it ,
1992. M ultiple-brooding and annual productivity of a
Neotropical m igrant passerine, The Black-throated
Blue W arbler (Dendroica caerulescens) in an unfrag­
m ented tem perate forest. Auk 109: 321 -333.
H o lm g re n , N., 1993. Patch selection, conflicting activities
and patterns o f migration in birds. Doctoral Disserta­
tion, Lund University.
H o u s to n , A.I. & J.M . M c N a m a r a , 1993. A theoretical in ve s­
tigation of the fa t reserves and m ortality levels of small
birds in winter. O m is Scand. 24: 2 0 5 -2 1 9 .
H o u s to n , A.I. & J.M . M c N a m a ra & J .M .C . H u tc h in s o n ,
1993. General results concerning the trade-off between
gaining energy and avoiding predation. Phil. Trans. R.
Soc. Lond. B 3 4 1 : 3 7 5 -3 9 7 .
H u n t in g f o r d , F., 1993. Behavioural m echanism s in evolu­
tionary perspective. Trends Ecol. Evol. 8: 81 -84.
Je n n i, L. & S. J e n n i-E ie rm a n n , 1992. M etabolic patterns of
feeding, overnight fasted and flying night m igrants d u r­
ing autum n migration. Ornis Scand. 23: 251 -259.
J e n n i-E ie rm a n n , S. & L. Je n n i, 1991. M etabolic responses
to flight and fasting in night-m igrating passerines. J.
Com p.Physiol. B 161: 4 6 5 -4 7 4 .
J e n n i-E ie rm a n n , S. & L. J e n n i, 19 92. High plasm a triglycer­
ide levels in sm all birds during m igratory flight: a new
pathway for fuel supply during endurance locom otion at
very high m ass-specific m etabolic rates? Physiol. Zool.
65: 1 1 2-12 3.
J e n n i-E ie rm a n n , S. & L. Je n n i, in press. Plasm a m etabolite
levels predict individual body-m ass changes in a small
long-distance m igrant, the Garden W arbler. Auk.
K a c e ln ik , A., J.R. K re b s & C. B e r n s te in , 1992. The ideal
free distribution m odel and predator-prey populations.
Trends Ecol. Evol. 7: 50 -55.
K e t t e r s o n , E.D. & v. N o l a n J r , 19 83. The evolution o f dif­
ferential bird migration. Current O rnithology I: 3 5 7 -4 0 1 .
K la a s s e n , M., in press. W ater and energy lim itation on flight
range. Auk.
K le ib e r , M., 1 9 6 1 . The fire of life. W iley, New York.
K lu ijv e r , H.N. & L. T in b e rg e n , 1953. Territory and the regu­
lation of density in titm ice. Arch. Néerl. Zool. 10: 2 6 6 287.
K r e b s , J.R., 1982. Territorial defence in the G reat T it (Parus
major): Do residents always win? Behav. Ecol. Sociobiol. 11: 18 5-19 4.
L e s s e lls , C.M., 19 91. The evolution of life histories. In: J.R.
K re b s & N.B. D avies. Behavioural Ecology, an e vo lu ­
tionary approach. Third edition, pp: 4 7 1 -4 8 8 . Blackwell
Scientific Publications, Oxford.
Le vin s, R., 1968. Evolution in changing environm ents: som e
theoretical explorations. Princeton Univ Press, Prince­
ton.
LlECHTl, F., 1992. Flugverhalten nächtlich ziehender Vögel
in A bhängigkeit von W ind und Topographie. PhD th e ­
sis, U niversity o f Basel.
— , 1993. N ächtlicher Vogelzug im Herbst über S üddeutsch­
land: W inddrift und Kom pensation. J. Orn. 134: 373404.
— , in press. Nocturnal m igrants adjusting air speed in rela­
tion to wind. Ibis 136.
L in d s t r ö m , Â., 1991. M aximum fat deposition rates in
m igrating birds. Ornis Scand 22: 12-19.
L in d s t r ö m , A . & T. A l e r s t a m , 1992. Optimal fat loads in
m igrating birds: a test of the tim e-m inim ization hypoth­
esis. A m . Nat. 140: 477-491.
L o r ia , D.E. & F.R. M o o r e , 1990. Energy dem and of m igra­
tion on red-eyed vireos, Vireo olivaceous. Beh. Ecol. 1:
24-35.
LUNDBERG, P., 1987. Partial bird m igration and evolutionary
stable strategies. J. theor. Biol. 125: 351-360.
LUNDBERG, S. & T. A l e r s t a m , 1986. Bird m igration patterns:
conditions for stable geographical population segrega­
tion. J. theor. Biol. 123: 403-414.
M a c A r t h u r , R.H., 1972. G eographical Ecology: patterns in
the distribution o f species. New York: Harper & Row.
M a d d o x , J,, 1993. Com petition and the death o f science.
Nature 363: 667.
M a n g e l , M. & C .W . C l a r k , 1986. Towards a unified fo ra g ­
ing theory. Ecology 67: 1127-1138.
M a n g e l , M. & C .W . C l a r k , 1988. Dynam ic modeling in
behavioural ecology. Princeton: Princeton University
Press.
M a r r a , P.P., T.W . S h e r r y & R.T. H o l m e s , 1993. Territorial
exclusion by older m ales in Neotropical m igrant bird in
winter: rem oval experim ents in Am erican Redstarts
(Setophaga ruticilla). Auk. 110: in press.
M a y n a r d S m it h , J „ 1982. Evolution and the theory of
games. Cam bridge University Press, Cambridge.
M c F a r l a n d , D .J. 1977. D e c is io n -m a k in g in a n im a ls . N atu re
269: 15 -21.
M c N a m a r a , J.M . & A .I. H o u s t o n , 1986. T h e c o m m o n c u r­
re n c y fo r b e h a v io u ra l d e c is io n s . A m . N at. 127: 358-
378.
M c N a m a r a , J.M. & A.I. H o u s t o n , 1992. State-dependent
life-history theory and its im plications for optim al clutch
size. Evol. Ecol. 6: 170-185.
M o o r e , F.R., 1992. Ecophysiological and behavioral
response to energy dem ands during migration. Acta
X X Congressus Internationalis Ornithologie: 753-760.
M o o r e , F.R., P. K e r l in g e r & T. S im o n s , 1990. Stopover on
a G ulf coast barrier island by spring trans-G ulf
m igrants. W ilson Bull. 102: 487-500.
M o o r e , F.R. & P.A. S im m , 1986. Risk-sensitive foraging by a
m igratory bird (D endroica coronata). Experientia 42:
1054-1056.
MYERS, J.P., 1981. A test of three hypotheses for latitudinal
segregation of the sexes in wintering birds. Can. J.
Zool. 59: 1527-1534.
O w e n , M. & J.M. B l a c k , 1989. Factors affecting the survival
o f barnacle geese on m igration from the breeding
grounds. J. Anim. Ecol. 57: 37-47.
P a p i , F. & H.G. W a l r a f f (eds), 1982. Avian Navigation. Ber­
lin: Springer.
P a r k e r , G .A ., 1974. A s s e s s m e n t s tra te g y an d th e e v o lu tio n
o f fig h tin g b e h a v io u r. J. th e o r. Biol. 47: 2 2 3 -2 4 3 .
P a r k e r , G.A. & W .J. S u t h e r la n d , 1986. Ideal free d istribu­
tions when individuals differ in com petitive ability: p he­
notype-lim ited ideal free models. Anim. Behav. 34:
1222-1242.
P e n n y c u ic k , C.J., 1969. The m echanics of bird migration.
Ibis 111: 525-556.
— , 1978. Fifteen testable predictions about bird flight. Oikos
30: 165-176.
— , 1989. Bird R ig h t Perform ance. Oxford: Oxford University
Press.
P ie n k o w s k i, M.W. & P.R. E v a n s , 1985. T h e ro le o f m ig ra ­
tio n in th e p o p u la tio n d y n a m ic s o f b ird s. In: R.M. S ib le
& R.H. S m ith . B e h a v io u ra l E co lo g y: e c o lo g ic a l c o n s e ­
q u e n c e s o f a d a p tiv e b e h a v io u r. O xfo rd : B la ckw e ll S c i­
e n tific P u b lica tio n s: 331-351.
P ie rs m a , T., 1987. Hop, skip or jum p? Constraints on m igra­
tion of arctic waders by feeding, fattening, and flight
speed. Lim osa 60: 185-191. [In Dutch with English
summ ary],
P ie rs m a , T., R. D r e n t & P. W ie rs m a , 1991. Tem perate ve r­
sus tropical wintering in the w orld's northernm ost
breeder, the knot: m etabolic scope and resource levels
restrict subspecific options. Acta XX Internat. Ornithol.
Congr. (Christchurch): 761-772.
P ie rs m a , T. & B.J. E n s, 1992. O ptim al m igration schedules:
reserve dynam ics as constraint. W ader Study Group
Bull. 64: 17-18.
P ie rs m a , T. & S. v a n de S a n t, 1992. Pattern and predicta­
bility of potential wind assistance for waders and geese
m igrating from W est Africa and the W adden Sea to
Siberia. Ornis Svecica 2: 55-66.
P ie rs m a , T., L. Z w a r t s & J.H. B ru g g e m a n , 1990. Behav­
ioural aspects of the departure of waders before long­
distance flights: flocking, vocalizations, flight paths and
diurnal tim ing. A rdea 78: 157-184.
P ip e r, W .H., 1990. Site tenacity and dom inance in wintering
white-throated sparrows Z onotrichia albicollis (Passeriform es: Em berizidae). Ethology 85: 114-122.
P ip e r, W .H. & R.H. W ile y , 1989. C orrelates of dom inance in
wintering white-throated sparrows: age, sex and lo ca ­
tion. Anim . Behav. 37: 298-310.
P ip e r, W .H . & R.H. W ile y , 1990. C orrelates of range size in
wintering white-throated sparrows, Zonotrichia albicol­
lis. Anim. Behav. 40: 545-552.
R a y n e r, J.M .V ., 1988. Form and function in avian flight.
Curr. Ornithol. 5: 1 -77.
S h e r r y , T.W . & R.T. H o lm e s , 1992. Population fluctuations
in a long-distance Neotropical m igrant: dem ogrpahic
evidence for the im portance of breeding season events
in the Am erican Redstart: 431-444. In: J.M. H a g a n III
a n d D.W. J o h n s to n . Ecology and C onservation of
Neotropical Migrant Landbirds (J.M. Hagan III and
D.W. Johnston, eds.). Sm ithsonian Press, W ash. D.C.
S h e r r y , T.W. & R.T. H o lm e s , in press. Sum m er versus w in­
ter lim itation of Neotropical m igrant landbird popula­
tions: w hat are the issues and w hat is the evidence? In:
T. M a r t in & D. F in c h . Ecology and M anagem ent of
N eotropical M igratory Birds: a Synthesis and Review of
the Critical Issues. O xford University Press, Oxford.
S h e r r y , T.W. & R.T. H o lm e s , M S. Patterns of winter habitat
use in Nearctic-Neotropical m igrant birds and their
im plications for population dynam ics and conservation.
S te a r n s , S .C ., 1992. The evolution o f life histories. Oxford
Univ. Press, Oxford.
S u t h e r la n d , W .J. & P.M. D o lm a n , in press. Com bining
behaviour and population dynam ics with applications
for predicting consequences of habitat loss. Proc Royal
Soc Series B.
S u t h e r la n d , W.J. & G .A. P a r k e r , 1985. Distribution of une­
qual com petitors. In: R.M. S ib ly & R.H. S m ith . Behav­
ioural Ecology: ecological consequences of adaptive
behaviour. Oxford: Blackwell Scientific Publications:
255-274.
T in b e rg e n , J.M . & S. D aan, 1990. Fam ily planning in the
great tit (Parus m ajoi): optim al clutch size as an inte­
gration o f parent and offspring fitness. Behaviour 114:
161-190.
T u c k e r , V .A ., 1973. Bird m etabolism during flight: evalua­
tion of a theory. J. Exp. Biol. 58 : 689-709.
WlERSMA, P. & T. PlERSMA, 1994. Effects o f m icrohabitat,
flocking, clim ate and m igratory goal on energy expend­
iture in the annual cycle o f red knots. Condor 9 6 : 257279.
Y app, W .B., 1962. Some physical lim itations in bird m igra­
tion. Ibis 104: 86-89.
Z w a r t s , L., B.J. E n s, M. K e r s te n & T. P ie rs m a , 1990.
Moult, m ass and flight range of waders ready to take
off for long-distance m igrations. Ardea 78: 339-364.
A P P E N D IX I: C O M P E TITIV E CO NCEPTS
D uring th e w o rksh o p and the fo llo w in g g e sta tio n o f th is re p o rt it be ca m e c le a r th a t m y d is lik e o f so m e term s
a nd co n c e p ts w a s n o t sha red by the m a jo rity o f pa rticip ants. S in ce m u d dled co n ce p ts w ill lead to m ud dled
in vestig ation s, I fe lt in spire d to w rite th is ap pe nd ix, w h ere I try to m ake cle a r on w h a t basis I distin g u ish
be tw ee n useful and co n fu sin g con cep ts. A c o n ce p t is con sid ered useful w h en it has a c le a r de fin itio n and, in
princip al, a llo w s ta kin g m ea su re m e n ts. A co n ce p t is con sid ered co n fu sin g w h e n it has b e en use d in m a n y d if­
fe re n t w ays, even th o u g h it w a s o rig in a lly c le a rly defined. A co n ce p t is a lso co n sid e re d co n fu sin g if it has no
cle a r m e a ning w h en a p plied to a real w orld situa tion , even tho ugh it m a y be p re cise ly d e fin e d in a m a th e m a ti­
cal m odel atte m p tin g to d e scrib e th a t real w o rld situation.
USEFUL CONCEPTS
social dominance
Aggregate measure of success in winning encounters at a given location against a specified set of
opponents, probably depending on both relative fighting abilities and payoffs of focal animal and oppo­
nents. It is often wrongly thought of, not as a relationship, but as the ability of an individual to dominate
others. This is due to the habit of calculating dominance scores without making reference to the oppo­
nents or to the location (e.g. Ens & Goss -Custard 1984).
fighting ability
T he a b ility o f an in dividu al to w in fig h ts w h en payoffs and po sitio nal a d va n ta g e s are c o n tro lle d for. It
sho uld not be co n fu se d w ith social dom inance, w h ich refers to a re la tio n sh ip and not a tru e a b ility o f an
in dividu al. T hou gh the co n ce p t is e a sy to grasp, it is ra th e r eva sive w h en it co m e s to actu al m e a s u re ­
m ent. T his is be cau se it is ve ry hard to control fo r all the c o n fo u n d in g factors.
payoff to winning a contest
Increase or reduction in the reproductive potential of an individual when it either wins or loses a contest;
should be measured as the change in Expected Future Reproductive Success (EFRS); see
(1987). Though quickly understood on paper, it is very hard to actually measure. Measurement inevita­
bly involves many additional assumptions.
it
foraging ability
T he a b ility o f an in dividu al to find food. It can be derived from w h a t
Grafen
Goss-Custard & Durell (1988)
have ca lle d the Inte rfe re n ce Free Intake R ate (IFIR) by co n tro llin g fo r prey d e nsity. IFIR is m e a su re d as
the rate o f fo o d in ta ke th a t an in dividu al ach ie ves in a certain lo catio n w h en it is n o t h in d e re d by oth er
in dividu als. T here cou ld w e ll be a tra d e -o ff be tw ee n forag in g a b ility and fig h tin g ability, sin ce de sig nin g
an a n im a l th a t is g o o d a t fig h tin g m ay co n flict w ith d e m a n d s th a t m ake the a n im a l g o o d a t e xp lo iting
resources. In fid d le r cra b s fe m a le s have tw o sm all cla w s fo r p icking up food, w h ile in m ales on e o f
the se fe e d in g cla w s is e n la rged to such an exte nt th a t it can no lo n g e r be used fo r fee d in g , b u t it is very
effective in fig h tin g w ith con spe cifics, as well as in attractin g fe m a le s
(Christy & Salmon 1984). It is m y
hunch th a t in m a n y species, m ales m ay be be tte r a t figh ting o v e r reso urce s and less e ffic ie n t a t e x p lo it­
ing th e se reso urce s a t lo w reso urce le vels co m p a re d to fem ales.
susceptibility to interference (STI)
The rate at w h ich foo d in ta ke de clin e s w h en the nu m b e r o f co m p e tito rs in crea ses, m o s t o fte n m e a s ­
(Goss-Custard &
Durell 1988). Ideally, it sho uld be co n tro lle d fo r prey density, b u t to m y kn o w le d g e th is has n e ve r been
ured as the slo pe o f th e re la tio n sh ip be tw ee n intake rate and co m p e tito r d e n sity
done.
CONCEPTS CAUSING CONFUSION
resource holding potential (RHP)
W hen P a r k e r (1974) laun ched his th e o ry th a t the o u tcom e o f e n co u n te rs w ill de p e n d on both th e fig h t­
ing a b ilitie s (e.g. w e a p o n ry o r physical streng th) and the pa yoffs to the in dividu als involved, he used the
w ord R e so u rce H olding Potential (RHP) instead o f figh ting ability. T hou gh in te nd ed as a cla rificatio n,
the term R H P has given rise to m uch m isu n d e rsta n d in g s in m y o p in ion and sho uld, perhaps, be
dro pp ed. A s fa r as I u n de rsta nd it, th e reason fo r a d op ting th e term RH P is th a t it also in clu d e s p o s i­
tion al a d van ta ge s. T w o kn ig h ts m ig h t b e n e fit e q u a lly from gaining acce ss to a p rince ss and m ig h t be
e q u a lly good figh ters, but if one ha s to fig h t up hill the od ds are a g a in st him . S im ilarly, a m ale to a d m ay
w ard o ff rival m a le s m ore e a sily w h en it is in a m p le xu s w ith the fe m a le tha n w h en he sw im s n e a r to her.
Thus, w h e n using the w o rd fig h tin g ab ility instead o f reso urce ho ld ing potential, it is im p o rta n t to keep in
m ind th a t the re could be positio nal a d van ta ge s, w h ich sho uld be un re la te d to the pa yoffs though.
competitive ability
M y princip al problem w ith the co n ce p t o f co m p e titive a b ility is th a t it crea te s th e im p re ssio n th a t it can
be m e a su re d as a sin g le ch a ra cte r o f an individual. I thin k th a t to be high ly unlikely. As used by S u t h ­
e r l a n d & P a r k e r (1985), co m p e titive a b ility in dica tes the a b ility to acq uire re so u rce s w h e n com p etin g
w ith o th e r in dividu als and it is m ore o r less eq ua te d to su sce p tib ility to in te rfe ren ce, i.e. th e rate at
w h ich foo d in ta ke de clin e s w hen th e nu m be r o f c o m p e tito rs in crea ses. S u sce p tib ility to in te rfe re n ce
d e pe nd s on local d o m in a n ce and it is ve ry likely th a t local d o m in a n ce d e pe nd s on both pa yoffs and
fig h tin g ability. W h e re a s figh ting a b ility is a tru e a b ility o f an individual, so cia l d o m in a n ce is not. W e
m ig h t d e cid e to e q u a te com p etitive a b ility to figh ting ability, but I th in k little w o uld be ga in ed from tw o
te rm s fo r a sin g le concept. F urtherm ore, the p o in t a b o u t c o m p e titive a b ility is th a t it d e scrib e s th e s u c ­
cess in com p etin g for resources. W hen scram b le co m p e titio n is th e m o st im p o rta n t process, fo ra g in g
ability, in ste ad o f figh ting ability, is w h a t m atters. W hen both co n te s t and scra m b le c o m p e titio n m atter,
w e need to m ea sure both forag in g a b ility and figh ting a b ility and the po ssib le tra d e o ff be tw ee n th e se
tw o a b ilitie s w o uld d e stro y a n y hope o f m ea surin g c o m p e titive a b ility as a sin g le cha racter. In fact, a
tra d e o ff see m s in evita ble to me, since the a ttrib u te s th a t m ake an an im a l g o od a t fig h tin g are d iffe re n t
from th e a ttrib u te s th a t m ake an an im a l good a t forag in g.
(w ritten by B runo Ens)
A P P E N D IX II: LIST O F PARTICIPANTS
S. Â k e s s o n , D e p a rtm e n t o f A nim al E cology, U n ive rsity o f Lund, S -2 2 3 62 Lund, S w eden.
P r o f . D r . T. A le r s t a m , D e p a rtm e n t o f A nim al E cology, U n ive rsity o f Lund, E kolog ihu set, S -2 2 3 62 Lund,
S w eden.
D r s . H. B a p t is t , R ijksw a te rsta at, DG W , P.O. Box 8039, 4 3 3 0 EA M iddelburg, T he N e th erland s.
D r . J.J. B e u k e m a , NIO Z, P.O. Box 59, 1790 A B Den Burg (Texel), T he N e th erland s.
D r . H. B ie b a c h , M ax P la nck In stitu t fü r V e rh a lte n sp h ysio lo g ie , D -8 13 8 A ndechs, G e rm a n y
D r s . D. B o s , R ijksw a te rsta at, DG W , P.O. Box 8039, 4 3 3 0 EA M iddelburg, T he N e th erland s.
T. B r e g n b a ll e , N ational E nviron m ental R e sea rch Institute, D e p ta rtm e n t o f W ild life E cology, Kalo, 8410
Ronde, D enm ark.
D r . B. B r in k m a n , Instituu t v o o r Bos- en N a tu u ro n d e rzo e k (IB N -D LO ), P.O. Box 167, 1790 A D Den Burg
(Texel), T h e N etherlands.
D r . B. B r u d e r e r , S ch w e ize risch e V og elw arte, C H -6 2 0 4 Sem pach, S w itze rlan d
D r s . J.H. B r u g g e m a n n , D e p a rtm e n t of M arine Biology, U n iversity o f G roningen, P.O. B ox 14, 9 7 5 0 A A Haren,
The N etherlands.
L. B r u in z e e l , Z o o lo g ica l La boratory, U n iversity o f G roningen, P.O. B ox 14, 9750 A A G ro n in g e n , T he N e th e r­
lands.
D r s . L.S. B u u r m a , L u ch tm a ch tsta f hoofd se ctie n a tu u rlijk m ilieu, P.O. B ox 20 703, 2 5 0 0 ES Den Haag, The
N e th erland s.
D r s . N. C a d é e , R W S d ie n s t g e tijd ew a te ren , afd eling biologie, P.O. B ox 8039, 4 3 3 0 EA M id de lbu rg, T h e N e th ­
erlands.
D r . R. C a l d o w , Institute o f T e rre stria l Ecology, F urzeb roo k R e sea rch S tation, W a reh am , D o rse t B H 2 0 5AS,
UK.
D r . D.A. C r is t o l , D e p a rtm e n t o f B iology, U n ive rsity o f Indiana, B lo o m in g to n IN 47405, Indiana, USA.
D r . N. D a n k e r s , Instituu t vo o r Bos- en N a tu u ro n d e rzo e k (IB N -D LO ), P.O. Box 167, 1 7 90 AD D en Burg
(Texel), T h e N etherlands.
A. D e k in g a , N IO Z, P.O. Box 59 1790 A B Den Burg The N etherlands.
S. D ie t r ic h , A m H e lge ha us 12, D -3 57 5 K irch ha in 7, G erm any.
P.M . D o l m a n , S cho ol o f B io lo gica l S ciences, U n iversity o f East A nglia, N orw ich, N o rfo lk N R 4 7TJ, UK.
D r s . M. v a n E e r d e n , RW S direktie F levoland, P.O. Box 600, 8 2 00 AP Lelystad, T he N e th erland s.
D r s . M. E n g e lm o e r , V a kg ro e p G ed ra gsb iolog ie, R ijksu n ive rsite it G roningen, P.O. B o x 14, 97 50 A A Haren,
The N etherlands.
D r . B.J. E n s , In stitu u t voo r B o s -e n N a tu uro nde rzoe k (IBN -D LO ), P.O. Box 167, 1790 A D Den Burg (Texel),
The N etherlands.
P r o f . D r . P.R. E v a n s , U n ive rsity o f D urham , D e p a rtm e n t o f B io lo gica l S ciences, D urham DH1 3LE, UK.
J.
van
G il s , V a kg ro e p G ed ra gsb iolog ie, R ijksu n ive rsite it G roningen, P.O. B ox 14, 9 7 5 0 A A H aren, T he N e th ­
erlands.
D r s . P. d e G o e ij, NIO Z, P.O. Box 59, 1790 A B Den Burg (Texel), The N etherlands.
D r . J.D. G o s s - C u s t a r d , In stitu te o f T e rrestria l Ecology, F urzeb roo k R esearch S tation, n e ar W a re h a m , D o rset
B H 205 AS, UK.
A. H e d e n s tr ö m , D e p a rtm e n t o f T h e o re tica l E cology, U n iversity o f Lund, E kolog ihu set, S -2 2 3 62 Lund, S w e ­
den.
P r o f . D r . R.T. H o l m e s , D e p a rtm e n t o f B iological S ciences, D a rtm o uth C ollege, H anover, N e w H a m psh ire
03 75 5-35 76 , USA.
D r . N. H o lm g r e n , D e p a rtm e n t o f T h e o re tica l E cology, E cology B uilding, U n ive rsity o f Lund, S -2 2 3 62 Lund,
S w eden.
D r . A.I. H o u s t o n , S cho ol o f B io lo gica l S ciences, W ood la nd Road, Bristol BS8 1UG, UK.
H. H u is m a n n , F achb ere ich B io lo gie /C he m ie, U n iversitä t O snabrück,
P.O. B ox 4469, 4 5 0 0 O sna brück, G e r­
m any.
D r . L. J e n n i , S ch w e ize risch e V og elw arte, C H -620 4 Sem pach, S w itze rlan d.
D r . S. J e n n i-E ie rm a n n , S ch w e ize risch e V og elw arte, C H -620 4 S em pach, S w itze rlan d.
M. K e s t e n h o lz , S ch w e ize risch e V ogelw arte, C H -620 4 S em pach, S w itze rlan d.
N. K j e l l é n , D e p a rtm e n t o f A nim al E cology, U n iversity o f Lund, S 2 2 3 62 Lund, S w eden.
D r . M. K l a a s s e n , M ax P la nck Institut fü r V e rh a lte n sp h ysio lo g ie , D -8 23 46 A nd ech s, G erm any.
A. K o o lh a a s , N IO Z P.O. B ox 5 9, 1 7 9 0 A B Den Burg (Texel), T h e N etherlands.
D r . F. L ie c h ti, S ch w e ize risch e V og elw arte, C H -6 2 0 4 S em pach, S w itze rlan d.
D r . Â . L in d s t r ö m , D e p a rtm e n t o f A nim al E cology, U n ive rsity o f Lund, E kologihuset, S -2 2 3 62 Lund, S w eden.
D r s . J. v a n d e r M e e r , NIO Z, P.O. Box 59, 1790 A B Den Burg (Texel), The N etherlands.
D r . C .D .T . M in t o n , R AO U , 21 G lad stone St., M oo ne e P onds, M elbourne, Vic. 3150, A ustralia.
I. M i t c h e l l , D e p a rtm e n t o f B io lo gica l S ciences, South Road, D urham DH1 3LE, UK.
T. M o o d e y , NER O U nit o f B ehavioural Ecology, D e p a rtm e n t o f Zoology, S ou th P arks Road, O xfo rd 0 X 1 3PS,
UK.
D r . F.R. M o o r e , D e p a rtm e n t o f B iological S ciences, U n ive rsity o f S o u thern M ississippi, B ox 5018, H a ttie s­
burg, M ississippi 39 40 6-50 18 , USA.
P r o f . D r . M .L. M o r t o n , D e p a rtm e n t o f Biology, O ccide ntal C ollege, Los A ng eles C A 90 04 1-33 92 , USA.
D r . A. v a n N o o r d w ijk , NIO O C e ntre fo r T errestria l Ecology, P.O. B ox 40, 66 66 ZG H eteren, T he N etherlands.
D r . M .E. P e r e y r a , D e p a rtm e n t o f Biology, O ccide ntal C ollege, Los A ng eles, C A 90 041-3392, USA.
D. P e t e r , S ch w e ize risch e V og elw arte, C H -620 4 S em p ach , Sw itzerland.
D r . T. P ie rs m a , NIO Z, P.O. Box 59, 1790 A B Den Burg (Texel), T h e N etherlands.
D r . B. P in s h o w , B la ustein Institute fo r D esert R esearch, S ed e B o q e r C am pus, 8499 3 Israel.
D r s . M. P o o t , NIO Z, P.O. Box 59, 1790 A B Den Burg (Texel), T he N etherlands.
J. S a lv ig , N ational E nviro n m e n ta l R esearch Institute, G renavej 12, Kalo, D K -8 41 0 Ronde, D enm ark.
D r . I. S c o t t , D e p a rtm e n t o f B iological S ciences, S outh Road, D urham DH1 3LE, UK.
H. S it t e r s , Institute o f T e rre stria l E cology, F urzeb roo k R esearch S tation, F urzeb roo k, W areh am , D orset
B H 2 0 5AS, UK.
D r s . C.J. S m it , Instituu t vo o r Bos- en N a tu u ro n d e rzo e k (IB N -D LO ), P.O. Box 167, 1790 A D Den B urg (Texel),
T he N etherlands.
R. S p a a r , S ch w e ize risch e V og elw arte, C H -6 2 0 4 Sem pach, S w itzerland.
H. S t a r k , S ch w e ize risch e V og elw arte, C H -620 4 S em p ach , Sw itzerland.
D r . J.M . T in b e r g e n , V a kg ro e p G ed ra gsb iolog ie, R ijksu n ive rsite it G roningen, P.O. B ox 14, 9 7 50 A A Haren,
T h e N etherlands.
D r s . I. T u l p , NIO Z, P.O. Box 59, 1790 A B Den Burg (Texel), T he N etherlands.
K. U h le n b u s c h , F achb ere ich B io lo gie /C he m ie, U n iversitä t O snabrück, P.O. Box 4 4 6 9 , 4 5 0 0 O sna brück, G e r­
m any.
A .J. U r f i , Institute o f T errestria l E cology, F u rze b ro o k R esearch S tation, F urzebrook, W a reh am , D o rse t B H 20
5AS, UK.
D r s . N. V e r b o v e n , N IO O -C TO , B ote rhoe ksestr. 22, P.O. B ox 40, 6666 ZG H eteren, T he N e th erland s.
D rs . Y. V e r k u il , NIO Z, P.O. B ox 59, 1790 A B Den Burg (Texel), T he N etherlands.
T.P. W e b e r , N E R C U nit o f B ehavioural E cology, D e p a rtm e n t o f Zoology, South P arks Road, O xford 0 X 1 3PS,
UK.
D r s . P. W ie rs m a , V a kg ro e p G ed ra gsb iolog ie, R ijksu n ive rsite it G roning en, P.O. B ox 14, 9 7 5 0 A A Haren, T he
N etherlands.
P r o f . D r . W .J. W o l f f , IB N -D LO , P.O. B ox 46, 3956 ZR Leersum , T h e N etherlands.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Reto Spaar
Jacob Saivig
Clive Minton
Dan Cristol
Theunis Piersma
Kerstin Uhlenbusch
Nils Kjéllen
Marcel Klaassen
M ennobart van Eerden
Bert Brinkman
Sabine Dietrich
Paul Dolman
Herbert Biebach
Yvonne Verkuil
Frank Moore
Bruno Bruderer
Karin Krijgsveld
Cor Smit
Thom as Bregnballe
Luit Buurma
Jan van der Kamp
Jaap van der Meer
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
Ian Mitchell
Martin Poot
Ian Scott
Maria Pereyra
Martin Morton
Sussanne Âkesson
Tony Moodey
Anita Koolhaas
Eric Stienen
Noël Holmgren
Paul Ruiters
Petra de Goeij
Niels Cadée
Meinte Engelm oer
Susi Jenni-Eierm ann
Alasdair Houston
Hans Schekkerman
Thomas W eber
Daan Bos
Âke Lindström
Humphrey Sitters
Lukas Jenni
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
Arie van Noordwijk
Bruno Ens
W im W olff
Richard Caldow
Dick Holmes
Heike Huism ann
Jan van Gils
John Goss-Custard
Leo Bruinzeel
Popko W iersm a
Felix Liechti
Ingrid Tulp
Dieter Peter
Peter Evans
Abdul Jamil Urfi
Herbert Stark
Henrich Bruggemann
Berry Pinshow
M athias Kestenholz
Joost Tinbergen
Thom as Alerstam
Anders Hedenström
CONTENTS
Preface......................................................................................................................................................................1
Towards predictive models of bird migration schedules: theoretical and empirical bottlenecks....................3
Introduction........................................................................................................................................................ 3
Defining the problem .........................................................................................................................................4
The migratory flig h t...........................................................................................................................................5
Flight mechanics..........................................................................................................................................5
Energy and other potentially limiting resources during m igration.......................................................... 6
The migratory journey: (re)fuelling, departing and flying.............................................................................. 8
Physiology of making and using nutrient stores...................................................................................... 8
Consistency in allometric relationships of physiological and ecological variables...............................9
Optimal departure fat loads........................................................................................................................9
Optimal return jou rn e y..............................................................................................................................10
Optimal decisions and stochasticity.........................................................................................................11
Measuring the fitness consequences of arrival date............................................................................. 12
Which models are best?.................................................................................................................................13
The full annual cycle as an individual decision............................................................................................13
Population dynamics of migratory species...................................................................................................14
Density-dependent reproduction..............................................................................................................14
Density-dependent winter survival.......................................................................................................... 15
The cost of m igration................................................................................................................................16
Predicting the effects of habitat change.......................................................................................................17
Conclusions................................... ................................................................................................................. 18
Some important and hopefully amenable problem s............................................................................. 18
Migratory birds provide good research systems....................................................................................18
Prospects...................................................................................................................................................18
References...................................................................................................................................................... 19
Appendix I: Competitive concepts...................................................................................................................... 23
Useful concepts..............................................................................................................................................23
Concepts causing confusion..........................................................................................................................24
Appendix II: List of participants...........................................................................................................................22