A USER MANUAL FOR SEAPODYM VERSION 2 . 0 : application with data assimilation ( draft

SEAPODYM is a model developed initially for investigating spatial tuna population dynamics under the influence of both fishing and environmental effects. The model is based on advection-diffusion-reaction equations. The main features of this model are i) forcing by environmental data (either observed or modeled), ii) prediction of both temporal and spatial distribution of age-structured populations, iii) prediction of total catch and size frequency of catch by fleet when fishing data (catch and effort) are available, iv) parameter optimization based on fishing data assimilation techniques. The model development also benefited of a grant from the PFRP (Pelagic Fisheries Research Program) of the University of Hawaii, allowing the implementation of irregular grids and initiating the work for parameter optimization (2004-05). Since 2006, the development has continued within the MEMMS section (Marine Ecosystem Modeling and Monitoring by Satellites) of the Spatial Oceanography Division of CLS, a subsidiary of the French CNES and IFREMER Institutes. Collaboration with SPC continues, with funding support from a new EU-funded SPC project, SCIFISH, as well as with the PFRP through a second grant (2006-09) for the project " 'Climate and Fishing Impacts on the Spatial Population Dynamics of Tunas (Project no. 657425) " '. An enhanced version of the model (SEAPODYM 2.0) is now ready for use. It includes revised mechanisms due to changes in the modelling of mid-trophic organisms of the pelagic ecosystem with several pelagic mid-trophic functional groups. Dynamics of tuna populations have been also revised with expanded definitions of habitat indices, movements, and natural mortality. The code for parameter optimization has been completed and tested. Following a general introduction and a brief review of the main mechanisms used in the model, this user's manual provides the information needed to run numerical simulations for tuna or tuna-like species, and the method to fit the model predictions to observations by estimating model parameters, investigation of parameter observability, improving initial conditions and the model functionality.