Re: Fish dispersal modelling and GIS

[M Lang / S Railsback]

address@hidden wrote:
> 
> It seems that this page has some information on your topic of interest. I
> haven´t read it through, but it seems to be in your line of interest.
> Greetings,
>      Konrad
> 
> http://math.humboldt.edu/~simsys/Products.html

Yes- that is our site and it is mostly about modeling fish in Swarm.
Look around. One page is about a Columbia River salmon simulator in
which we import a field of points where water velocity is known in two
dimensions and the fish swim and drift in the current field. This
simulator is now in the custody of Jim Anderson at U. Washington and I'm
sure he would send you the model description.

In that salmon simulator we imported the current data from a
hydrodynamic model. For each point in horizontal space where the
hydrodynamic model simulates velocity (called a 'node'), we imported the
point's X and Y coordinates, the X and Y coordinates of the current
velocity at the point, and the depth. When it was time for an individual
fish to move, the fish looked around to find the nearest node and used
the current velocity from that node. If the nodes were spaced on a
regular grid, the fish would not have to do this looking around, it
could calculate exactly where the nearest node is. 

Many years ago there was a simulator of ocean salmon movement, again I
think the fish moved via a combination of swimming and drifting with the
current. It was called NerkaSim; there is still probably a web page up
for it, but I don't think the project is active. It was written in
VisualBasic.

Paul Box at Utah State probably has more code than anyone else for
moving GIS data in and out of Swarm. You should check out how he does
it. 

We've done several models similar to various parts of what you are
considering, and I'm sure you could make it work in Swarm. We started a
model of queen conch but did not get to the larvae dispersal phase.
However, we did develop a graphical interface that could be useful to
you- there is one display of the whole "world" showing the density of
fish in each grid cell; then you can click on a grid, tell it how many
grids you want to zoom in on, and open up a new display showing the
locations of the individuals. Let me know if you want it.

Feel free to email me directly if you want more info on anything at our
web site. If you are just learning how to model fish growth, etc. take a
look at the full description of our stream trout model.

Steve
 
> 
> Hello,
> 
> Here at York University (England) we are  currently working on an
> individual based model of fish larval dispersal running  under behavioural
> and ecological constraints.  I am trying to figure out  how to integrate
> the model to GIS and hopefully generate a model of tidal and  oceanic
> current movements outside or inside the GIS as a source of environmental
> information and forcing for the dispersal model.  I am very interested
> if anyone may have used swarm methods or any agent-based  modelling with
> fish or even plankton being affected by currents or  insects in wind maybe,
> do you think swarm may be a good base for building models  of currents and
> wind, with agents moving around in this mobile  environment??
> I also would be interested in the code you use to handle  data between
> models and data from ArcGIS grids so deploying Arc/Info or ArcView  as a
> 'data file manager' and display tool.  How much work exactly would you
> envisage GIS doing where flow could be modelled?
> 
> Our fish will be  growing over a 1 month period, developing from a
> passively dispersed propagule  into a seeing, hearing and smelling fish
> with preferences to where it will  settle, and a strong swimming ability.
> 
> If anyone has any opinions  on this I look forward to hearing your reply,
> also if anyone knows of any  relevant reading material I would be very
> grateful for the details.


-- 
address@hidden
Lang, Railsback & Assoc.
250 California Ave., Arcata CA 95521
707-822-0453; Fax 822-1868


                  ==================================
   Swarm-Modelling is for discussion of Simulation and Modelling techniques
   esp. using Swarm.  For list administration needs (esp. [un]subscribing),
   please send a message to <address@hidden> with "help" in the
   body of the message.
                  ==================================