Space use by passerine birds : a study of territory economics in robins Erithacus rubecula and dippers Cinclus cinclus
1. Cost constraints in models of territory size are based on time/activity/laboratory estimates that predict birds using larger territories will incur higher energy costs. The predicted form of the cost constraint may be linear, accelerating or decelerating depending on assumptions inherent in the models. The aim of this study was to assess the reality and form of the cost constraint by making direct measurements of the energy costs of territory use in birds that occupy territories of different size and shape; polygonal territories represented by the robin Erithacus rubecula, and linear by the dipper Cinclus cinclus. Free-living energy expenditure was measured using the doubly-labelled water technique, whilst simultaneously recording patterns of territory use by radio-tracking. 2. Territorial robins concentrated their activity in one or more foraging patches located in bushes. Range polygons containing all the foraging patches used by an individual provided estimates of territory area, and were generally of high eccentricity. A small proportion of robins was classified as non-territorial based on range polygon areas. Furthermore, while territorial robins showed high fidelity to ranges over the short term (days), non-territorial individuals were nomadic. Over the longer term (months), however, some territorial robins showed range drift. Dippers similarly used preferred core regions within ranges, although there was no selection for particular habitat features. 3. Because robins occupied territory polygons which varied from polygonal to highly linear, work was focused on this species to allow intra-specific comparison. Robins tended to commute between foraging patches by flying. It was appropriate, therefore, to describe territories in terms of a number of patches linked by a network of flight paths. This generated two further measures of territory size; the number of patches used and the total flight distance between patches. 4. The robins exploited a renewing food supply. Predictions were tested concerning the temporal scheduling of visits to foraging patches within territories. Patches tended to be separated by flight paths of similar lengths, and were visited in a regular sequence. Although the number of foraging patches used varied, all territories had similar total core areas. Robins using many small foraging patches commuted between patches more often and covered a larger total flight distance during each foraging circuit of the territory. The configurations of foraging patches were used in a highly linear manner. This was true even if the territory containing them was of low eccentricity. 5. Changes in structure and pattern of use varied predictably with territory size, and could be described mathematically. Based on this and published time/activity budgets, a suite of models was developed to predict how energy costs would vary with number of patches used and total flight distance between patches. Models were tested by directly measuring the energy expenditure of robins using different territories. The number of patches used and total flight distance between patches were both significantly correlated with energy expenditure, while territory area was not. One of the models showed a significant fit to the observed data, and suggested that the form of the energy cost constraint on territory size was linear. The effect of territory shape on energy costs was minimal. The implications of these results for models of territory size are discussed. 6. The slope and elevation of the energy cost constraint varied with the morphology of territory occupants. Based on this, an association of morphology with territory size was predicted; robins of lower mass and wing-loading using larger territories. The observed data supported these predictions, and suggested a possible genetic predisposition to particular patterns of territory occupancy in the robin.