Reducing model complexity via output sensitivity

Ecosystem models help us understand the mechanisms that influence ecosystem health indicators. However, if they are too complex, these mechanisms can be difficult to identify. On the other hand, if they are too simple the mechanisms may be distorted or even absent. Determining an appropriate level of model complexity is therefore desirable. This paper introduces two model simplification methods that are based on the sensitivity of performance measures to model rates and components. The first method identifies rates that have little influence on the performance measures and subsequently eliminates them. The second identifies, for a given performance measure, state variables that can be made constant. The methods can be implemented automatically, so that familiarity with the model is not required a priori. Demonstrating with a biogeochemical model of Port Phillip Bay, Australia, we find that significant reduction in model complexity is possible, including reductions in model order. Also, the process of implementing the methods reveals insights into the system that were not obvious beforehand.

[1]  Sašo Džeroski,et al.  Constructing a library of domain knowledge for automated modelling of aquatic ecosystems , 2006 .

[2]  Hironori Hirata,et al.  Modelling and analysis of ecological systems: the large-scale system viewpoint , 1987 .

[3]  Elizabeth A. Fulton,et al.  The effects of model structure and complexity on the behaviour and performance of marine ecosystem models , 2001 .

[4]  Saso Dzeroski,et al.  Integrating Knowledge-Driven and Data-Driven Approaches to Modeling , 2006, EnviroInfo.

[5]  J. R. Kramer,et al.  Evaluating modelling uncertainty for model selection , 2001 .

[6]  Karl-Erich Lindenschmidt,et al.  The effect of complexity on parameter sensitivity and model uncertainty in river water quality modelling , 2006 .

[7]  Robert B. McKane,et al.  Belowground processes in forest-ecosystem biogeochemical simulation models , 2000 .

[8]  I. Dor,et al.  Hypertrophic Reservoirs for Wastewater Storage and Reuse , 1999 .

[9]  Y. Svirezhev,et al.  Aggregation of variables in simulation models of water ecosystems , 1983 .

[10]  Paul-Gerhard Martin,et al.  The use of canonical S-system modelling for condensation of complex dynamic models , 1997 .

[11]  D. Fletcher,et al.  Equivalence relationships between stage-structured population models. , 2002, Mathematical biosciences.

[12]  Jock Lawrie,et al.  A METHOD FOR SIMPLIFYING LARGE ECOSYSTEM MODELS , 2008 .

[13]  Roger L Breton,et al.  A comparison of model performance for six quantitative structure‐activity relationship packages that predict acute toxicity to fish , 2003, Environmental toxicology and chemistry.

[14]  D. Sorensen,et al.  A Survey of Model Reduction Methods for Large-Scale Systems , 2000 .

[15]  Sašo Džeroski,et al.  Learning population dynamics models from data and domain knowledge , 2003 .

[16]  Pat Langley,et al.  Constructing explanatory process models from biological data and knowledge , 2006, Artif. Intell. Medicine.

[17]  A. Duncan,et al.  Modelling the ecological aspects of bankside reservoirs and implications for management , 1999, Hydrobiologia.

[18]  Sven Erik Jørgensen,et al.  Improved calibration of a eutrophication model by use of the size variation due to succession , 2002 .

[19]  H. Bugmann.  A Review of Forest Gap Models , 2001 .

[20]  P. Ruardij,et al.  The European regional seas ecosystem model, a complex marine ecosystem model , 1995 .

[21]  Graeme E. Batley,et al.  Port Phillip Bay Environmental Study Final Report , 1996 .

[22]  Robert V. O'Neill,et al.  Aggregation error in ecological models , 1979 .

[23]  Jeremy S. Collie,et al.  Are multispecies models an improvement on single-species models for measuring fishing impacts on marine ecosystems? , 2000 .

[24]  S. Walker.  Hydrodynamic models of Port Phillip Bay , 1997 .

[25]  W. G. Cale,et al.  Behavior of aggregate state variables in ecosystem models , 1980 .

[26]  S. Sheppard,et al.  A soil solute transport model evaluated on two experimental systems , 1987 .

[27]  A. Wood,et al.  Towards the systematic simplification of mechanistic models , 2006 .

[28]  M. B. Beck.  Hard or soft environmental systems , 1981 .

[29]  Marten Scheffer,et al.  A strategy to improve the contribution of complex simulation models to ecological theory , 2005 .