Cargando…
Post-Hoc Pattern-Oriented Testing and Tuning of an Existing Large Model: Lessons from the Field Vole
Pattern-oriented modeling (POM) is a general strategy for modeling complex systems. In POM, multiple patterns observed at different scales and hierarchical levels are used to optimize model structure, to test and select sub-models of key processes, and for calibration. So far, POM has been used for...
Autores principales: | , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Public Library of Science
2012
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3457952/ https://www.ncbi.nlm.nih.gov/pubmed/23049882 http://dx.doi.org/10.1371/journal.pone.0045872 |
Sumario: | Pattern-oriented modeling (POM) is a general strategy for modeling complex systems. In POM, multiple patterns observed at different scales and hierarchical levels are used to optimize model structure, to test and select sub-models of key processes, and for calibration. So far, POM has been used for developing new models and for models of low to moderate complexity. It remains unclear, though, whether the basic idea of POM to utilize multiple patterns, could also be used to test and possibly develop existing and established models of high complexity. Here, we use POM to test, calibrate, and further develop an existing agent-based model of the field vole (Microtus agrestis), which was developed and tested within the ALMaSS framework. This framework is complex because it includes a high-resolution representation of the landscape and its dynamics, of the individual’s behavior, and of the interaction between landscape and individual behavior. Results of fitting to the range of patterns chosen were generally very good, but the procedure required to achieve this was long and complicated. To obtain good correspondence between model and the real world it was often necessary to model the real world environment closely. We therefore conclude that post-hoc POM is a useful and viable way to test a highly complex simulation model, but also warn against the dangers of over-fitting to real world patterns that lack details in their explanatory driving factors. To overcome some of these obstacles we suggest the adoption of open-science and open-source approaches to ecological simulation modeling. |
---|