You are here

“It was an artefact not the result”: A note on systems dynamic model development tools

Publication Type:

Journal Article


Environmental Modelling & Software, Volume 20, Issue 12, p.1543-1548 (2005)


Lotka-Volterra equation; Predator-prey modelling, MATEMATICA, MATLAB, Model analysis, MODELMAKER, Numerical ODE solvers, Simile, STELLA, Stiff systems


<p>Environmental modelling is done more and more by practising ecologists rather than computer scientists or mathematicians. This is because there is a broad spectrum of development tools available that allows graphical coding of complex models of dynamic systems and help to abstract from the mathematical issues of the modelled system and the related numerical problems for estimating solutions. In this contribution, we study how different modelling tools treat a test system, a highly non-linear predator&ndash;prey model, and how the numerical solutions vary. We can show that solutions (a) differ if different development tools are chosen but the same numerical procedure is selected; (b) depend on undocumented implementation details; (c) vary even for the same tool but for different versions; and (d) are generated but with no notifications on numerical problems even if these could be identified. We conclude that improved documentation of numeric methods used in the modelling software is essential to make sure that process based models formulated in terms of these modelling packages do not become &ldquo;black box&rdquo; models due to uncertainty in integration methods.</p>


<p><a name="SeppeltRicter2005News">07 October: Integration methods test model added to the model catalogue, </a><a href="../../../../../../examples/catalogue/modeldescription.php?Id=predator_prey32.sml">Lotka-Volterra predator-prey model (Seppelt and Richter, 2005)</a>. <br />
The model is included in response to the call by <a href="">Seppelt and Richter (2005)</a> for evaluation of the numerical methods of simulation modelling tools. Such evaluation is to be welcomed as it is required to provide confidence in results, though of course, the model itself must also represent the target system adequately for confidence. If either the model representation or numerical methods are flawed, the results will be flawed. Unfortunately, Seppelt and Richter (2005) report aberrant behaviour of Simile 3.2 though neither the authors nor we can reproduce the flawed behaviour. A <a href="">corrigendum</a> has been published.</p>