[R-sig-dyn-mod] Individual-based modelling

[Thomas Petzoldt]

Hi Sibylle,

a non-spatial individual based model can be found in the following JSS 
paper (**):

https://www.jstatsoft.org/article/view/v022i09

... that is also a package vignette of simecol. The class 
"indbasedModel" is now built-in.


Instead of the empirical equations found in "main" of the model above, 
you may consider to replace (or amend) this with an ODE model that 
describes the live cycle of an **individual**, e.g. a bioenergetic model.

This means that the ODE part of the model does NOT describe the 
population size, because this is done by the IBM part counting the 
population. The population size is a so-called "emergent property" of 
the model.

Package simeolModels 
(https://r-forge.r-project.org/scm/viewvc.php/pkg/simecolModels/?root=simecol)

contains an ODE empowered IBM. You can install it via:

  install.packages("simecolModels", repos="http://R-Forge.R-project.org")

and then have a look at ?daphnia_deb_ibm


Note that all these are just examples, and that many other approaches 
are possible (e.g. with spatial coordinates), depending on your needs.

Implementing individual-based model in R can be much fun for an 
experienced R person, because it is very flexible. Compared to pure ODEs 
it requires a little bit more programming skills, so it would be a good 
idea to do it in a team or on supervision of an experienced person. Even 
more important is a clear idea what to do and of course enough 
creativity to think around the corners.

I don't yet understand your goals in detail, but would suggest to start 
with a simple non-spatial IBM along the lines of the JSS paper. Here the 
population is represented as a data frame and its columns are the traits 
of the individuals. When this works, start to refine the model.


Hope it helps,

Thomas

(**) Another implementation of a Daphnia IBM without the "simecol" 
framework was already published in R News 03(2003) 
http://cran.r-project.org/doc/Rnews/Rnews_2003-3.pdf



-- 
Dr. Thomas Petzoldt
Technische Universitaet Dresden
Faculty of Environmental Sciences
Institute of Hydrobiology
01062 Dresden, Germany

E-Mail: thomas.petzoldt at tu-dresden.de
http://tu-dresden.de/Members/thomas.petzoldt

-- limnology and ecological modelling --