Hi all,
I have been tasked with looking at the possibility of incorporating a foodweb ecosystem model into ROMS, specifically looking at krill populations. This would include predator species, fishing and fishery management rules.
I am new to ocean physics/ecology and modelling in general, being just a lowly programmer by trade.
Does anyone have suggestions on the practicality of doing something like this? Does it sound feasible? Would it be better to incorporate it into ROMS or to create a separate model and couple it to ROMS using MCT or ESMF?
I am not yet sure whether there would be any feedback from this to the ocean model, possibly just one way data flow from ocean -> foodweb model?
I would like to hear any ideas for where to start on this project, what past work to look through, or whether it is just not worth pursuing.
Thanks
Troy
Foodweb model
Re: Foodweb model
What is currently in ROMS is the lower trophic level models. These have nutrients, phytoplankton, zooplankton. Krill are a sort of zooplankton, right? I don't know if any NPZ models in ROMS are tuned for southern ocean krill. The typical NPZ model has idealized losses on the highest trophic level, just some general mortality that's usually a function of their density.
Modeling the predators and fishing fleets is a whole different game, involving a different group of experts. There are IBM (individual based models) for various fish/crab/etc. species which are often run from the stored physics and NPZ fields providing behavior to the individuals, but the coupling back to the Z is quite a challenge and very much a field of active research. It's a lot to ask of a "lowly programmer".
Honestly, if you got this to work, you'd come out with a paper having 10-20 authors because it would take that many people working together on the programming, the fleet behavior, the whale behavior, getting the krill to seem like krill, including realistic interannual variability in the temperature, ice, etc, etc.
Modeling the predators and fishing fleets is a whole different game, involving a different group of experts. There are IBM (individual based models) for various fish/crab/etc. species which are often run from the stored physics and NPZ fields providing behavior to the individuals, but the coupling back to the Z is quite a challenge and very much a field of active research. It's a lot to ask of a "lowly programmer".
Honestly, if you got this to work, you'd come out with a paper having 10-20 authors because it would take that many people working together on the programming, the fleet behavior, the whale behavior, getting the krill to seem like krill, including realistic interannual variability in the temperature, ice, etc, etc.