Ecological Modelling 177 (2004) 95–106 Modelling the effects of different quality prey fish species and of food supply reduction on growth performance of Roseate Tern chicks Irene Martins a,∗ , José Costa Pereira a , Jaime A. Ramos a , Sven Erik Jørgensen b a IMAR—Institute of Marine Research, c/o Department of Zoology, University of Coimbra, 3004-517 Coimbra, Portugal b DFH, Environmental Chemistry, University Park 2, Copenhagen Ø, Denmark Received 19 August 2003; received in revised form 27 January 2004; accepted 16 February 2004 Abstract We assessed the effects of different quality fish species and reductions in their abundance in the growth of Roseate Tern (Sterna dougallii) chicks in the Azores. An energetic balance model was developed, which accounts for the energy available from ingested fishes and the energy expenditure by the chicks. Fish species ingested by Roseate Tern chicks were mainly Trachurus picturatus, Scomberesox saurus, Macroramphosus scolopax and Capros aper. Energy expenditure was defined by the basal metabolic rate (BMR), which depends on chick’s weight, plus a multiple of BMR, obtained by calibration and accounting for unspecified energetic losses. Real data obtained in June–July of 1996 and 1995 were used to calibrate and validate the model, respectively. Regressions between observed and predicted data showed a higher fitting level for 1996 than for 1995, which seems to be related with the need for more accurate estimations of assimilation efficiency and energy expenditure parameters. Of the four main prey species, T. picturatus was the most energetically favourable because of its high length–weight regression and to the capacity of chicks to swallow T. picturatus with an average length of 8.71 cm. C. aper had a higher length–weight relationship, but chicks were only able to ingest items <5 cm because this species is wider than T. picturatus. S. saurus and M. scolopax were energetically less efficient because they had a lower length–weight relationship. Observations suggest that the availability of less energetic fish (e.g. M. scolopax) leads to higher delivery rates by the parents and, consequently, to higher ingestion rates by the chicks. However, both observations and predictions suggest that the growth efficiency is lower than when the chicks were fed with T. picturatus with an average length of 8.71 cm. Additionally, the model predicts that chicks with 23 days of age will attain 85 and 56 g, according to a 25% reduction in prey within a year of energetically more and less favourable prey, respectively. Chicks in the first situation are still likely to fledge, while in the second case, chick survival is certainly compromised. In summary, the growth of Roseate Tern chicks in the Azores will be significantly reduced in response to a decrease in energy intake resulting from provisioning of low quality fish species and chick survival is compromised whenever this factor is associated with a significant (25%) reduction in fish stocks. This model proved to be a highly dynamic tool in assessing variations in postnatal growth of Roseate Terns in response to variations in food quality and quantity and it should be of interest in the conservation strategy of this species. © 2004 Elsevier B.V. All rights reserved. Keywords: Energetic balance model; Chick growth; Sterna dougallii; Trachurus picturatus; Fish stock reductions; Azores ∗ Corresponding author. Tel.: +351-239-834729; fax: +351-239-823603. E-mail address: imartins@ci.uc.pt (I. Martins). 0304-3800/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.ecolmodel.2004.02.002