Five software platforms for scientific agent-based models (ABMs) were reviewed by implementing example models in each. NetLogo is the highest-level platform, pro-viding a simple yet powerful programming language, built-in graphical interfaces, and comprehensive documentation. It is designed primarily for ABMs of mobile individuals with local interactions in a grid space, but not necessarily clumsy for others. NetL-ogo is highly recommended, even for prototyping complex models. MASON, Repast, and Swarm are “framework and library ” platforms, providing a conceptual framework for organizing and designing ABMs and corresponding software libraries. MASON is least mature and designed with execution speed a high priority. The Objective-C version of Swarm is the most mature library platform and is stable and well-organized. Objective-C seems more natural than Java for ABMs but weak error-handling and the lack of developer tools are drawbacks. Java Swarm allows Swarm’s Objective-C li-braries to be called from Java; it does not seem to combine advantages of the two languages well. Repast provides Swarm-like functions in a Java library and is a good choice for many, but parts of its organization and design could be improved. A rough comparison of execution speed found MASON and Repast usually fastest (MASON 1-35 % faster than Repast), Swarm (including Objective-C) fastest for simple models but slowest for complex ones, and NetLogo intermediate. Recommendations include com-pleting the documentation (for all platforms except NetLogo), strengthening concep-tual frameworks, providing better tools for statistical output and automating simulation experiments, simplifying common tasks, and researching technologies for understand-ing how simulation results arise.

Abstract—Model design is particularly challenging for multiagent simulation models as the simulation paradigm does hardly impose constraints on it. This contribution systematically analyzes procedures for developing a multi-agent simulation model: iterative methods derived from principles, such as KISS or KIDS and methods focussing on the different design elements (agents, interaction, environment). I.

Individual-based modelling and analysis of forest experiments has been made more accessible to researchers with the advent of modelling environments like Simile from www.simulistics.com. Individual-based analyses of tree growth data offer insights not possible with plot-based analyses, especially when the original experimental design has been compromised by mortality or other unforeseen events. The paper illustrates how Simile can be used for individual-based analyses of mixed plantings, and how it can be used to explore the consequences of the resulting statistical models. A mixed-species planting of Eucalyptus pellita and Acacia peregrina is used to illustrate possibilities.

The simecol package provides an open structure to implement, simulate and share ecological models. A generalized object-oriented architecture improves readability and potential code re-use of models and makes simecol-models freely extendable and simple to use. The simecol package was implemented in the S4 class system of the programming language R. Reference applications, e.g. predator-prey models or grid models are provided which can be used as a starting point for own developments. Compact example applications and the complete code of an individual-based model of the water flea Daphnia document the efficient usage of simecol for various purposes in ecological modeling, e.g. scenario analysis, stochastic simulations and individual based population dynamics. Ecologists are encouraged to exploit the abilities of simecol to structure their work and to use R and object-oriented programming as a suitable medium for the distribution and share of ecological modeling code. Note: A previous version of this introduction to the R package simecol has been published as Petzoldt and Rinke (2007) in the Journal of Statistical Software,

This paper addresses problems in computational creative discovery, either autonomous or in synergetic tandem with humans. A computer program generates output as a combination of base primitives whose interpretation must lie outside the program itself. Concepts of combinatoric and creative emergence are analysed in relation to creative outputs being novel and appropriate combinations of base primitives, with the conclusion that the choice of the generative process that builds and combines the primitives is of high importance. The generalised concept of an artificial ecosystem, which adapts concepts and processes from a biological ecosystem at a metaphoric level, is an appropriate generative system for creative discovery. The fundamental properties of artificial ecosystems are discussed and examples given in two different creative problem domains. Systems are implemented as pure simulation, and where the ecosystem concept is expanded to include real environments and people as ecosystem components, offer an alternative to the ‘software tool ’ approach of conventional creative software. Keywords: Emergence. Artificial ecosystems, Combinationalism, “Theories are important and indispensable because without them we could not orientate ourselves in the world — we could not live. Even our observations are interpreted with their help.” — Karl Popper, The Myth of the Framework 1

Abstract not found

This paper summarises the history of the terms ecology and ecosystem, before examining their application in the early and recent literature of A-Life agent-based software simulation. It investigates trends in A-Life that have led to a predominance of simulations incorporating artificial evolution acting on generic agents, but lacking a level of detail that would allow the emergence of phenomena relating to the transfer and transformation of energy and matter between the virtual abiotic environment and biota. Implications of these characteristics for the relevance of A-Life’s virtual ecosystem models to Ecology are discussed. We argue a position that the inclusion of low-level representations of energetics, matter and evolution, in concert with pattern-oriented modelling techniques from Ecology for model validation, will improve the relevance of A-Life models to Ecology. We also suggest two methods that may allows us to meet this goal: artificial evolution can be employed as a mechanism for automating pattern-oriented ecological modelling from the level of individual species up to that of the ecosystem, or it may be employed to explore general principles of ecosystem behaviour over evolutionary time periods.

Abstract not found

Abstract not found

Bacterial communities such as biofilms are widely recognised as being important for survival and persistence of bacteria in harsh environments. Mechanistic models of biofilm growth indicate that the way in which the surface is seeded can effect the morphology of simulated biofilms. Experimental studies indicate that genes which are important for chemotaxis also influence biofilm formation, perhaps by influencing aggregation on a surface. Understanding aggregation and microcolony formation could therefore help clarify factors influencing biofilm formation as well as understanding how groups may influence the fitness of bacteria. In this paper I develop an Individual Based Model to examine how different behaviours involved in microcolony formation on a surface determine patterns of group sizes, and link patterns to bacterial fitness. I also provide a method for comparing data with model hypotheses to identify bacterial behaviours in experimental systems. 1