AUTHOR=Tareen Samar H., Ahmad Jamil , Roux Olivier TITLE=Parametric linear hybrid automata for complex environmental systems modeling JOURNAL=Frontiers in Environmental Science VOLUME=3 YEAR=2015 URL=https://www.frontiersin.org/journals/environmental-science/articles/10.3389/fenvs.2015.00047 DOI=10.3389/fenvs.2015.00047 ISSN=2296-665X ABSTRACT=

Environmental systems, whether they be weather patterns or predator–prey relationships, are dependent on a number different variables, each directly or indirectly affecting the system at large. Since not all of these factors are known, these systems take on non-linear dynamics, making it difficult to accurately predict meaningful behavioral trends far into the future. However, such dynamics do not warrant complete ignorance of different efforts to understand and model close approximations of these systems. Toward this end, we have applied a logical modeling approach to model and analyze the behavioral trends and systematic trajectories that these systems exhibit without delving into their quantification. This approach, formalized by René Thomas for discrete logical modeling of Biological Regulatory Networks (BRNs) and further extended in our previous studies as parametric biological linear hybrid automata (Bio-LHA), has been previously employed for the analyses of different molecular regulatory interactions occurring across various cells and microbial species. As relationships between different interacting components of a system can be simplified as positive or negative influences, we can employ the Bio-LHA framework to represent different components of the environmental system as positive or negative feedbacks. In the present study, we highlight the benefits of hybrid (discrete and continuous combined) modeling which lead to refinements among the fore-casted behaviors in order to find out which ones are actually possible. We have taken two case studies: an interaction of three microbial species in a freshwater pond, and a more complex atmospheric system, to show the applications of the Bio-LHA methodology for the timed hybrid modeling of environmental systems. Results show that the approach using the Bio-LHA is a viable method for behavioral modeling of complex environmental systems by finding timing constraints while keeping the complexity of the model at a minimum.