Research Interests

Behavioural adaptations, proximate mechanisms, and constraints in reproduction and parental investment in northern bird populations



A. Tit studies

Ph.D., docent Seppo Rytkönen
Department of Biology
University of Oulu


See also:

B. Escape distance studies








Tit studies in the project "Behavioural adaptations, proximate mechanisms, and constraints in reproduction and parental investment in nortnern bird populations"

Objectives & Short description.

The study is based on the finding that there are differences in reproductive success among northern Parus species. The most striking difference is between the willow tit P. montanus and the great tit P. major, a local species with high breeding success and a newcomer with poor success. The main questions are whether northern tits are food limited during the breeding period, whether this is due to mismatch between timing of breeding and local resource abundance peak (which are more variable than in Central Europe), and how all this affects the parental investment decisions. Long-term data on timing of breeding, breeding success and food abundance are analysed to find out the mechanisms behind the reproductive performance of the tits. Besides decisions of timing and clutch size, the parental decisions crucial for succeeding in the North include breeding habitat, foraging niche, and food selection. However, it is probable that all these could not be optimised because of ecomorphological, physiological, phylogenetic and environmental (e.g. climatic changes) constraints. The results of this project will increase our knowledge of the possible effects of global warming on northern nature.



Objectives and Methods





A1. Backgrounds of the Parus-project

There is some evidence that variation in abundance and timing of the most important food of breeding passerines is greater in northern Europe than in Central Europe (Veistola 1997, own observations). This, however, might not be problematic for breeders that have adopted the clutch adjustment strategy (Pettifor et al. 1988), if the birds have proper behavioural adaptations and decision rules for predicting the forthcoming food resources. In central latitudes great tits and blue tits (Gibb 1950) and willow tits in higher latitudes have clutch adjustment strategy, whereas great tits in higher latitudes seem to have some kind of brood reduction strategy (Orell 1983, Sasvari & Orell 1992).

Tits form a phylogenetically and ecologically close group, which shares similar diet and foraging niche during the breeding season (Perrins 1991). We expect that they have basically similar adaptations for breeding and parental care in the similar environmental conditions. Many tit species live in mosaic environments and/or geographically varying conditions, which could be solved in two ways: to specialise in certain environmental conditions and search for these, or to evolve phenotypic plasticity to cope with different conditions (e.g. Dias & Blondel 1996). By studying the behavioural adaptations for breeding of willow, great and blue tits in northern Europe, we can get insight to the roles of phenotypic plasticity, specialisation and maladaptation in northern bird populations.


A2. Objectives and Methods


Food-limitation during the nestling period: an extra-feeding experiment

Diet and foraging niche in relation to spatio-temporal changes in food resources

Parental investment in relation to variation in food resources



Food-limitation during the nestling period: an extra-feeding experiment

Food is the most important proximate factor influencing reproduction in birds (Lack 1947). Parents that raise their offspring when food resources are most plentiful are the most successful breeders (Perrins 1991, van Noordwijk et al. 1995). Many studies have tested the hypothesis that food availability is the proximate factor affecting the timing of breeding and clutch size (for review, see Nager et al. 1997). However, few studies have experimentally investigated the role of food availability in the nestling period (Verhults 1994). In tits, which have adopted the clutch adjustment strategy, food limitation during the nestling period could reveal a local maladaptation in this strategy.

Hypothesis and predictions: Locally adapted birds that can adjust their clutch size to the natural food resources should not be food-limited during the nestling period. We expect that the willow tit is not food-limited in its main population area in northern Europe. However, great and blue tits, which are newcomers and represent "southern decision rules" in northern Europe, should be food-limited if they cannot match their breeding to the magnitude and/or timing of the caterpillar peak.

Tests and methods: Extra-fed broods of great, blue and willow tits are compared with control ones. Breeding tits are offered living Calliiphora larvae, by which parents can feed their young (Calliphora larvae has been proved to be accepted as nestling food in willow and great tits). It is easily possible to satiate about 1/4 to 1/3 of the maximum energy demand of the young with Calliphora larvae diet (energy content is about 7 kJ per gram fresh mass). During the extra-feeding experiment the seasonal changes in natural food resources (particularly in caterpillar abundance) are monitored with frass-fall method and beating trays (Sutherland 1997). The outcome of the study is related to the parent’s ability to adjust the timing of breeding and clutch size to the natural food resources.


Diet and foraging niche in relation to spatio-temporal changes in food resources

The Central European tits succeed best in deciduous (oak and beech) forests, where caterpillars are the most important food resource. Selection works for the synchronisation of breeding with caterpillar growth using temperature as a predictor (van Noordwijk et al. 1995). In northern Europe, however, there is large annual variation in the proportion of caterpillars in the diet (in willow tits: 20-80%, unpublished data), and the importance of other prey species (spiders, dipteras, aphids, etc.) is considerable (Rytkönen et al. 1996). There is also spatial variation. Our preliminary data show that there are considerable differences in the caterpillar abundance between the tree species. The peak of Epirrita autumnata in birches is always the first peak of the season. The caterpillar abundance in pines and spruces increases later in the season, while the caterpillars in birches can reduce dramatically. Due to mosaic landscape, these tree-specific differences result in differences between the breeding habitats. These findings raise a question whether the ability to broaden the diet and/or foraging niche when caterpillars are scarce is the key to the success in locally adapted tits?

Hypothesis and predictions: We suggest that the ability to broaden diet width and foraging niche when food is less abundant is an important adaptation in varying northern conditions. Thus, we expect that the willow tit has broader diet and a more diverse foraging niche than great (and perhaps blue) tits, especially when the breeding does not match the caterpillar peak.

Tests and methods: Foraging niches (and niche widths) and diets of great, blue and willow tits are compared to the availability of resources in different stages of breeding season. Foraging niches are determined by following foraging birds during the breeding season and quantifying "tree-use" (Lahti et al. 1998). Food items the birds consume are studied by 1) observing foraging birds and identifying their prey during the course of breeding season, 2) collecting/video monitoring the food loads which parents bring to their nestlings. Prey species are determined mainly to family level, which better enables geographical comparisons. Food availability is determined in niches where tits forage: 1) on ground level by pitfall-method, 2) in ground vegetation and bushes by sweep nets, 3) in trees by beating trays and frass-fall collectors.


Parental investment in relation to variation in food resources

Iteroparious organisms (i.e. those breeding more than once) are assumed to invest in their offspring according to the expected net benefits of the behaviour, which takes into account the survival prospects of the current brood and the chances of future reproduction (e.g. Rytkönen et al. 1990). Primary investment is the clutch size, and during the course of the breeding cycle, parental behaviour is expected to reflect the net reproductive value of the offspring. General predictions for parental investment would be that parental expenditure and willingness to take risks in nest defence should increase with offspring age, number and quality. Brood survival prospects, i.e. brood value, is undoubtedly dependent on the availability of food resources during the nestling period.

Hypothesis and predictions: Broods that have the advantage of perfect timing (matching the caterpillar peak) would be more valuable for the parents, and therefore, the optimal amount of parental investment in them would be greater than that in broods laid too early or too late. Therefore, we should find parents taking greater risks in nest defence and putting more effort in nestling provisioning in nests where the timing match the caterpillar peak, and in broods which have offered extra-food during the nestling period.

Tests and methods. Nest defence and nestling provisioning effort are measured in broods differing in natural or experimental food resources. Sex-specific parental provisioning behaviour is recorded by using video monitoring. The method is described in Rytkönen et al. (1995b). Parental anti-predatory responses against a predator model (a stoat Mustela erminea) are recorded. Risk-taking is assumed to correlate negatively with parental approach distances, positively with alarm calling rate and number of movements around the predator model. The method is described in Rytkönen et al. (1995b). See above for the methods of extra-feeding and monitoring the natural food resources.


A3. Results

The results of the studies dealing with long-term data on timing of breeding, timing of food abundance peaks, and breeding success in relation to the previous would give insight to the future developement in the northern nature, escpecially since the predictions of the climatic change due to global warming are quite dramatic in the northern Europe.

Results will be published in international Journals and presented in national and international congresses.

The following manuscripts are under construction:


A4. References

Dias, P.C. & Blondel, J. 1996: Local specialization and maladaptation in the Mediterranean blue tit (Parus caeruleus). – Oecologia 107: 79-86.

Gibb, J. 1950: The breeding biology of the great and blue titmice. – Ibis 92: 507-539.

Lack, D. 1968: Ecological Adaptations for Breeding in Birds. Methuen, London.

Nager, R.D., Rüegger, C. & van Noordwijk, A.J. 1997: Nutrient or energy limitation on egg formation: a feeding experiment in great tits. – J. Anim. Ecol. 66: 495-507.

Orell, M. 1983: Breeding and mortality in the great tit Parus major and the willow tit P. montanus in Oulu, northern Finland. – Acta Univ. Oul. A 148.

Perrins, C.M. 1991: Tits and their caterpillar food supply. – Ibis 133 (Suppl.): 49-54.

Pettifor, R.A., Perrins, C.M. & McCleery, R.H. 1988: Individual optimisation of clutch size in great tits. - Nature 336: 160-162.

Rytkönen, S. Koivula, K. & Orell, M. 1990: Temporal increase in nest defence intensity of the willow tit (Parus montanus): parental investment or methodological artifact? -Behav. Ecol. Sociobiol. 27:283-286.

Rytkönen, S. Koivula, K. & Orell, M. 1996: Patterns of per-brood and per-offspring provisioning efforts in the Willow Tit Parus montanus. - J. Anim. Biol. 27: 21-30.

Rytkönen, S. Orell, M., Koivula, K. & Soppela, M.1995: Correlation between two components of parental investment: nest defence intensity and nestling provisioning effort of willow tits. - Oecologia 104: 386-393.

Sasvári, L. & Orell, M. 1992: Breeding success in a North and a Central European population of the Great Tit Parus major. Ornis Scand. 23: 96-100.

Sutherland, W. J. (ed.) 1996: Ecological Census Techniques: A Handbook. CUP.

Van Noordwijk, A.J., McCleery, R.H. & Perrins, C.M. 1995: Selection for the synchronisation of great tit (Parus major) breeding with caterpillar growth, using temperature as a predictor. - J. Anim. Ecol. 64: 451-458.

Veistola, S. 1997: The effects of food and weather conditions on the breeding of hole-nesting passerines in the north. - Annales Universitas Turkuensis, Ser. A. II - TOM. 94.

Verhulst, S. 1994: Supplementary food in the nestling phase affects reproductive success in pied flycatchers (Ficedula hypoleuca). - Auk 111: 713-716.


Escape distance studies in the project "Behavioural adaptations, proximate mechanisms, and constraints in reproduction and parental investment in nortnern bird populations"

Second project tries to understand the variation in escape distances of breeding birds, especially those with precocial species (e.g. shorebirds, ducks and tetraonids). It is shown that the escape distance is an important behavioural trait that affects nest survival (mainly through predation). The aim is to develop an analytic model for optimal escape distance for breeding birds, and to test this model with empirical data in nature and with computer simulations. In addition, parental investment theory is applied to the parental escape behaviour, which can be considered as decision making in the parental investment behaviour.



Objectives and Methods









B1. Backgrounds of the escape distance project

Avian reproductive system combined with parental care composes a unique problem for an incubating or brooding bird parent. How to respond to threat of predation when a part of the inclusive fitness can be taken with when escaping (the parent itself) and a part must be left in the nest (eggs or nestlings). My interest about escape distances raised when I realised the striking difference in escaping behaviour between feeding waders and waders that attend their nests – even in the same habitats. Thus, the behaviour of the escaping birds "must" show certain anti-predatory adaptations. At least the pressure for anti-predatory adaptations is high, since predation rates of ground nesting birds are very high (e.g. Lack 1968).

Earlier findings. Earlier ideas about the significance of avian escape distances base on two findings: some breeding birds tend to leave the nest as early as they notice the predator and some birds sit very tightly in the nest. Byrkjedal (1987) proposed the following explanation. 1) Escaping bird would suffer the lowest nest predation rate when leaving the nest as early as it is possible in the environment. By doing so predator might not detect the nest (and escape) at all. This predation probability follows a detectability function which is a monotonically decreasing function of the distance from the predator to the nest (see e.g. Burnham et al. 1980). Thus, shorter escape distances would result in higher nest predation probability. 2) However, if the predator approaches closer than the critical distance for early departure, parent bird should remain tightly on the nest and not start performing distraction displays until the predator approaches certain threshold distance, which is close to the nest. According to Byrkjedal (1987), these two factors explain the bipolarity of escape distances (see also Gochfeld 1984): early departure would be the best strategy, and the advantages of sitting-tight strategy are especially related to distraction displays.

My idea. The problem in the above thinking is that the nest predation rate was studied in relation to the predator’s distance from nest. My point of view is to think escaping distance as a parental strategy, and the problem is, 1) at which distance the chances of nest survival would be the highest, or more generally 2) at which distance the parent’s inclusive fitness would be the highest?

The basic model. My model is based on three steps of nest predation as a consequence of parent bird’s escape from a nest. The probability of these steps depend on escape distance (d) as follows:

          1. Fescape(d): The frequency of escapes increase with increasing escape distance
          2. Pescape(d): The probability that predator detects the escape decreases with increasing escape distance.
          3. Pnest(d): The probability that the predator detects the nest after having detected the escape decrease with increasing escape distance

The relative nest predation probability as a function of escape distance (d) would be the product of these three steps:

Relative Pnest predation(d) = Fescape(d) * Pescape(d) * Pnest(d)

The relevance of the model was successfully "tested" in spring 1998, when the staff of the Department of Theoretical Ecology from University of Lund visited Oulu. I presented the idea in a meeting at Hailuoto Biological Station. The idea that predation probability (as a function of escape distance) is a product of the above three steps was accepted. Thus, the basics of the model are OK. Roger Härdling promised to co-operate in this project.

Optimal escape strategy. The adaptive significance of escape distance is based on an idea that predators get information of the location of the nest from the parent bird’s escape. Thus, the chances for the predator to find the nest can be divided into two factors, which depend on parent’s behaviour: A) chances to find the escape (# 1 & 2 above), and B) chances to find the nest by using the escape information (# 3). The predator’s chances are the best when the parent bird’s escape strategy is a mid-distance escape: escapes are rather frequent, predator’s chances to detect these are rather good, and the probability to eventually find the nest is rather high. The strategy of the parent bird is to reduce the above chances of the predator. The basic model predicts two alternative strategies for the best nest survival.

          1. As-short-as-possible escape distances would minimise the number of escapes the predator could observe (# 1).
          2. As-long-as-possible escape distance would decrease the predator’s chance to observe the escape (though escapes are frequent) (# 2), and minimise the predator’s chance to find the nest after the predator has observed the escape (# 3).

Additional factors.

These factors should also be included into the model of optimal escape distance.


B2. Objectives and Methods

The aim of the project is to develop an analytic model for optimal escape distance for breeding birds, and to test this model with empirical data in nature and with computer simulations.

I have some field observations of the escape distances of waders (gathered in Ulkokrunni, near Oulu). Additional data can be gathered on the meadows at Liminganlahti and Tauvo (large wetland and shore meadow areas nearby Oulu). Waders Charadriiformes are perhaps the most suitable group of birds in testing the model, since they are precocial (critical period is incubation, when the detectability of the nest is constant), lay equal-sized clutches (normally 4 eggs), and are abundant in different habitats. Since escape distance is a measure of parental investment, it can be affected by renesting chances (i.e. chances to start another breeding attempt after having lost the first attempt) in the population. Therefore, it is also important to study populations which differ in renesting potential. This can be done by studying populations in different geographical locations - the renesting potential decreases towards the North. This opens good possibilities to co-operate with other wader researchers all over the World - first connection would be Ingvar Byrkjedal in University of Bergen, Norway.

In co-operation with dr. Pekka Helle from Finnish Game and Fisheries Institute (GRI), I have planned to collect escape distances of different tetraonid species Galliformes. Pekka Helle has already some data from year 1998, and it is possible to recruit co-operative and well-motivated hunters in data collection (in case of tetraonids).


The study of escape distances in the field

Nests of the studied species are located in the field. Data can be easily gathered from the following waders: Common Sandpiper Actitits hypoleucos, Redshank Tringa totanus, Common Snipe Gallinago gallinago, Ringed Plover Charadrius hiaticula, Oystercathcer Haematopus ostralegus, Ruff Philomachus pugnax, Curlew Numenius arquata, Lapwing Vanellus vanellus; and the following tetraonids: Capercaillie Tetrao urogallus, Black Grouse Tetrao tetrix, Hazel Grouse Bonasa bonasia, Willow Grouse Lagopus lagopus.

The stage of breeding cycle is determined from the relative weight of the eggs (eggs lose water, and thus weight, when incubated). When the nest is located, the habitat structure is determined, as well as the surrounding of the nest cup (height and density of vegetation, upper cover, visibility to different directions).

If possible, the escape distance is measured at the first visit. However, the actual escape distance data is collected while visiting the known nests - this is the only way to get objective escape distances, since the chances to find the first escape depends on the parent bird’s escape distance (as the model predicts!). At each visit air temperature, wind velocity, vegetation height (plants grow...), other birds’ alarm calls or other activities, time of day, etc. are recorded.

Computer simulations

The adaptive significance of escape distance of a hypothetical bird can be surveyed by using computer simulations. The present version of the program "allows" a predator to search for a nest of a breeding bird that has a certain escape distance. The "size" or detectability of the nest can be varied, as well as the predator’s searching behaviour: how precisely it can locate the nest, and how intensely it searches for the nest after it has detected the escape? Later versions will include other factors possibly affecting the nest predation probability or the inclusive fitness of the parent (see above).


B3. Results

The model of optimal escape distance under construction is a new idea for science.

There are several possible applications for the escape distances:

Results will be published in international Journals and presented in national and international congresses.

The following manuscripts are under construction


B4. References

Burnham, K.P., Anderson, D.R. & Laake J.L. 1980: Estimation of density from line transect sampling of biological populations. – Wildl. Monogr. 72.

Byrkjedal, I. 1987: Antipredator behavior and breeding success in Great Golden-Plover and Eurasian Dotterel. – Condor 89: 40-47.

Gochfeld, M. 1984: Antipredator behavior: Aggressive and distraction displays of shorebirds, p. 289-377. In: J. Burger & B. Olla (eds.) Behavior of marine animals, Vol. 5. Shorebirds: breeding behavior and popultions. Plenum Press, New York.

Lack, D. 1968: Ecological adaptations for breeding in birds. Methuen, London.