Klaus Regelmann

Determinants of brood defence in the great tit Parus major L.

SummaryGreat tits (Parus major) tending nestlings reacted defensively to a live predator (Glaucidium perlatum; domestic cat) and the playback of a mixed species mobbing chorus, or to the latter alone. Defensive behaviour, mainly mobbing, reflected the risk taken and is assessed by five measures. Multivariate and contingency analyses revealed that at least 11 of 16 contextual independent variables affected the risk taken. Incremental effects are due to: Age of young, sex of the defending bird, the expected number of neighbouring mobbers, low temperature, wet canopy, the raptors distance from cover, coniferous forest, advancing season. A decremental effect is exerted by a large brood that is older. Annual differences in defence arise probably from demographic factors such as fecundity, which in turn affect the parents benefit-cost ratio (number of young of the same sex as the parent/residual reproductive value of the parent).While the effects of annual fecundity, age of young and season were predicted on the basis of this benefit-cost ratio, the failure to verify an incremental effect of brood size runs counter to established theory. We conclude that parents gear their defence efforts to energy investment, past or future, and are mal-adapted to brood size as a promotor of risk taken. The influence of the habitat is poorly understood. At least three factors (age and number of young, parents sex) act additively on part of the response. Despite the large number of variables examined, about 43% of the total response variance remains unexplained.While four defence measures are determined by at least 10 contextual factors, a fifth measure, the males minimum distance from the raptor, is determined by one other factor, the appearance of the ♀ male. The latter leads us to assume an additional, social rôle of brood defence.Risk-assessment by great tits leading to risk-aversive defence behaviour is governed by evolved restraints rather than by momentary constraints. Examples are provided by the effects of weather and cover.

Competitive resource sharing: A simulation model

Abstract The problem of how competing individuals should distribute themselves between food resource patches has been studied theoretically and experimentally. In this study a simple simulation model is used as a tool. To give the model a realistic background it is assumed that the individuals differ in their abilities to compete for food. Simulations are run with and without assuming travelling costs in order to study their influence. It is shown that the individuals distribute themselves between the food patches in good approximation to the ratio of the patch profitabilities. This result is discussed in relation to the theories of the ideal free distribution and the despotic distribution. The model makes five predictions on how competing individuals should distribute themselves between food resource patches. These predictions have already received some confirmation in experimental studies.

Fortpflanzungswert und „Brutwert“ der Kohlmeise(Parus major)

Untersuchungen zum Risikoeinsatz der Brutverteidigung durch Kohlmeisen bewogen uns, den Fortpflanzungswert vt (nachFisher) abhängig von Alter, Geschlecht, Biotop und Jahresbrutenzahl zu ermitteln (Abb. 1–3). Obwohl in die Berechnungen von vt viele Annahmen verschiedener Sicherheitsgrade eingehen, stellen die vt-Kurven die hinsichtlich der verarbeiteten demographischen Größen vollständigsten und daher genauesten uns bisher bekannter dar. Wir überprüften den vt-Verlauf mit Hilfe der Netto-Reproduktionsrate R0 unter der Annahme einer langfristig stabilen Populationsgröße und fanden ihn für einen Nadelwald vorzüglich bestätigt, für Laub(misch)wälder möglicherweise überschätzt (Tab. 3). ♀ haben stets ein niedrigeres vt als ♂, ausgenommen 1jährige ♂; unter diesen haben nur die Brüter ein höheres vt. Aus den vt-Kurven ermittelten wir zu den Ausfliegezeitpunkten ein Nutzen-Kosten-Verhältnis (ausfliegende Junge mt je Elter/Rest-Fortpflanzungswert vRt) abhängig von Brutbiotop, Alter, Geschlecht und Jahresbrut (Abb. 4, 5). Dieses Maß sollte dem Elter zum Prüfstein für die Stärke seines Bruteinsatzes dienen. Aus Betrag und Verlauf des Maßes werden fünf Vorhersagen zum Bruteinsatz von Kohlmeisen hergeleitet; die am wenigsten erwartete besagt, daß einbrütige Kohlmeisen im Laubwald denselben Einsatz zeigen sollten wie im Nadelwald, obwohl sie andere Fortpflanzungsraten (mt) haben. Da der Verlauf von vt wie von vRt während eines Brutzyklus nicht näher bekannt ist, schufen wir das neue Maß des „Brutwertes“. Er gibt den Wert der Brut für den Elter vom Nestbau bis zum Ausfliegen unter Berücksichtigung der Sterblichkeit von ♂, ♀ und Brut sowie der geschlechtsspezifischen Brutpflegefähigkeit der Elternvögel an (Abb. 7). Der Brutwert wächst von 0 (Nestbeginn) bis 100 % (Ausfliegen) an und ist dann identisch mit mt. ♀ haben anders als bei vt, nur während eines bestimmten Teils des Brutzyklus einen höheren Brutwert als die ♂ und sollten dann mehr an Kosten für die Brut aufwenden. Es wird erörtert, weshalb weder der geschlechtsspezifische Verlauf von vt noch jener vom Brutwert den höheren Einsatz des ♂ bei der Feindabwehr vom Nest erklären. Studies of the risk taken by great tit parents while defending their brood prompted us to computeFishers reproductive value vt as a function of age, sex, breeding habitat, and number of broods per year (Fig. 1–3). Although marred by the speculative nature of some underlying assumptions the resulting vt curves are most accurate for any animal in terms of their demographic comprehensiveness. They were checked via the net reproductive rate Ro assuming long-term stability of population numbers. The demographic ingredients on which vt was based proved to be in excellent agreement with stability (R0=1) for a larch forest but may have been less well estimated for deciduous woods (Table 3). ♀ vt values are always lower than ♂ vt values with the exception of yearling ♂ which only when breeding exceed ♀ values throughout. vt values at fledging were used to compute a cost benefit ratio of parental effort (number of fledglings mt per parent/residual reproductive value vRt) as a function of breeding habitat, age, sex and of whether a first or second brood was at stake (Fig. 4, 5). This measure is expected to provide a yardstick determining the efficiency of parental care during breeding. The cost benefit ratio thus defined is known most precisely at fledging and is thought to deviate from this value as one works backward in time up to the start of a breeding cycle when it is zero. The amount and the change in time of the cost benefit ratio lead to five non-trivial predictions about parental effort by great tits; the least obvious says that great tits producing one brood per year should invest the same effort regardless of type of habitat (deciduous vs coniferous wood) despite their reproductive rate (mt) differing greatly between habitats. Since the precise course of both vt and vRt during a breeding cycle are unknown we conceived of a new measure termed “brood value”. It reflects the value of the brood for each parent separately from the start of a nest up to fledging; the “brood value” takes into account the mortalities of the ♂, the ♀, and of the brood as well as the sex-specific potential for parental care (Fig. 7). Brood value increases from 0 (start of nest) to 100 % (fledging) when it becomes identical with mt. Brood value of ♀ is higher than that of ♂ only during part of the cycle as compared to vt. During this phase ♀ should invest more into their brood than ♂. It is discussed why neither the sex-specific course of vt nor that of brood value predict a higher male effort of brood defense as actually observed.

Learning to forage in a variable environment

In a stochastic environment animals should and really do make their foraging decisions not only according to the expected net benefits but also according to the associated variances. In this study I show how the RPS (relative payoff sum)-learning rule ( Harley, 1981 ) deals with the problem of choosing between two food patches where the expected net benefits are equal but the associated variances differ. The RPS-learning rule is a stochastic rule in which the probability of choosing a patch is proportionate to the relative payoff that the respective patch has yielded so far. To account for a decay of memory in time the RPS-rule gives more weight to recent payoffs. By using two different simulation procedures I show that the RPS-learning rule is either indifferent between constant and variable rewards or develops a preference for constant over variable rewards depending on the foraging regime simulated. The RPS-learning rule therefore represents a mechanism to produce risk-neutral or risk-averse foraging preferences. It is discussed how this result could come about and what it means in relation to empirical findings in a number of animal species.