INTER-AMERICAN WORKSHOP ON ENVIRONMENTAL DATA ACCESS

openModeller - a framework for biological/environmental modelling

Mauro E S Muñoz (), Renato De Giovanni (), Sidnei de Souza ()

Centro de Referência em Informação Ambiental

Abstract:

The project's goal is to build an open source framework to help users build species distribution models using different modelling algorithms and to help researchers develop and test such algorithms.

Currently there are several mathematical, statistical and computational algorithms for species distribution model generation. Examples are GARP, GAM, GLM and Bioclim. There are also new knowledge areas that could be used to generate such algorithms like cellular automata, fuzzy logic, neural nets and cognitive agents.

Despite their differences, all algorithms for species distribution modelling share some computational infrastructure needs. All must be able to:
  • read georeferenced environmental maps stored in different formats (GeoTiff, Arc/Info Grid, GXF, etc)
  • deal with different coordinate systems and projections to combine the different maps and the species occurrence points
  • resample the environmental characteristic maps and the species occurrence points
  • generate the species distribution map based on the resulting model.
The openModeller framework implements these general algorithms' needs and lets the user choose what environmental layers, species occurrence points and algorithm will be used to generate the resulting model.

On the other hand, openModeller lets the researcher run his/her own algorithm and compare the result with other algorithms with the certainty that they run exactly under the same conditions.

Another feature is that openModeller is a core multiplatform application, which means that developers can build their own graphical or web interface communicating with openModeller through C++ library calls.

This project is an initiative of the Reference Center on Environmental Information, CRIA (http://www.cria.org.br) and is been funded by FAPESP (http://www.fapesp.br).

* Keywords: spatial distribution model, species distribution model.

Organization:

Sponsors: