Data inputs for the African Penguin Spheniscus demersus model to be coupled to the pelagic OMP

Plagányi, Éva E; Robinson, William

Data inputs for the African Penguin Spheniscus demersus model to be coupled to the pelagic OMP

Working Paper

2008

Publisher

University of Cape Town

Department

Marine Resource Assessment and Management Group

Faculty

Faculty of Science

Abstract

This document serves as an updated compilation of all data currently available as inputs to the African penguin spatial model which is to be coupled to the pelagic OMP. The data are presented here together with some comments as to how they are to be used in the model and notes on their derivation and potential reliability. The model presented thus far is spatial in that different populations of penguins are represented, and different levels of movement between these populations are modelled. The main focus of the model is on Dassen and Robben Islands, which were originally combined for reasons of simplicity and because of their close proximity to each other, suggesting that the effects of external factors such as food availability would be highly correlated between the two. However, data that have recently become available indicate differences between these two colonies which suggest that it may no longer be appropriate to pool the two; hence they are split in the model. The third population is Dyer Island because it has the next largest numbers of penguins, recent declines in the population there are of concern and it is considered an important breeding site for penguins given the eastward shift of sardines. The fourth population is Boulders. Although relatively small, this colony was considered important to include because of its position, its role as the focus of several other studies and because penguins are known to have moved from Dyer Island to Boulders, Robben and Dassen, and hence it is useful to quantify to what extent movement of birds away from Dyer Island could account for observed declines at Dyer and increases at these other colonies. A summary of all the breeding colonies of penguins in so-called area i) is provided in Fig. 1 which also shows the relative abundance of breeding pairs in the different sub-areas, computed from data in Underhill et al. (2006). The regional penguin population is dominated (in terms of numbers) by two large colonies, namely Robben Island and Dassen Island; thus the model here has focused on these two colonies, with the next most important colony being Dyer Island. Fig. 2 maps the extent of strata corresponding to pelagic fish biomass estimates used to link to penguin breeding success in the model. Initially relationships were investigated with the west of Cape Agulhas pelagic spawner biomass and the west of Cape Infanta recruit abundances rather than the total South African pelagic fish abundance. More recently, this has been refined further still to use the Cape Columbine to Cape Point spawner biomass component only since this more accurately depicts the biomass available to penguins in the west coast model area. The west of Cape Infanta recruit estimates are retained because the anchovy and sardine recruits move down the West Coast. The model time step is one year and hence average trends are modelled. Penguins in each subarea are modelled starting from 1986.