Automated discovery of a model for dinoflagellate dynamics

The aim of this paper is to discover a model equation for predicting the concentration of the algal species Peridinium gatunense (Dinoflagellate) in Lake Kinneret. This is a rather difficult task, due to the sudden ecosystem changes that occurred in the mid-1990s. Namely, the stable ecosystem (with regular Peridinium blooms until 1993) underwent changes and has transformed into an unstable system, with cyanobacterial blooms now occurring regularly. This shift in the algal succession is expected to influence attempts to model the lake ecosystem. Namely, the model structure before and after the change is likely to be different. Our modelling experiments were directed to discover a single model equation that can simulate dinoflagellate dynamics in both periods. We apply an automated modelling tool (Lagramge), which integrates the knowledge- and the data-driven modelling approach. In addition we include an expert visual estimation of the models discovered by Lagramge to assist in the selection of the optimal model. The dataset used included time-series measurements of typical data from the periods 1988 to 1992 and 1997 to 1999. Using the data and expert knowledge coded in a modelling knowledge library, Lagramge successfully discovered several suitable mathematical models for Peridinium. After the expert's visual estimation and validation of the models, we propose one optimal model capable of long-term predictions.

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

[2]  Zvy Dubinsky,et al.  CO2 availability, carbonic anhydrase, and the annual dinoflagellate bloom in Lake Kinneret , 1994 .

[3]  Ljupco Todorovski.  Inductive Process Modeling , 2010, Encyclopedia of Machine Learning.

[4]  Wendy D. Welsh.  Water balance modelling in Bowen, Queensland, and the ten iterative steps in model development and evaluation , 2008, Environ. Model. Softw..

[5]  Niels Nyholm,et al.  A Simulation Model for Phytoplankton Growth and Nutrient Cycling in Eutrophic, Shallow Lakes , 1978 .

[6]  N. M. J. Crout,et al.  Is my model too complex? Evaluating model formulation using model reduction , 2009, Environ. Model. Softw..

[7]  Jürgen Benndorf,et al.  A Contribution to the Phosphorus Loading Concept , 1979 .

[8]  Matthew R. Hipsey,et al.  Implementation of ecological modeling as an effective management and investigation tool: Lake Kinneret as a case study , 2009 .

[9]  Ljupčo Todorovski,et al.  Using domain knowledge for automated modelling of dynamic systems with equation discovery , 2003 .

[10]  Richard A. Park,et al.  Documentation of selected constructs and parameter values in the aquatic model cleaner , 1976 .

[11]  Friedrich Recknagel,et al.  Computational Assemblage of Ordinary Differential Equations for Chlorophyll-a Using a Lake Process Equation Library and Measured Data of Lake Kasumigaura , 2006 .

[12]  Tamar Zohary,et al.  Changes to the phytoplankton assemblage of Lake Kinneret after decades of a predictable, repetitive pattern , 2004 .

[13]  T. Berman,et al.  PHOSPHORUS, NITROGEN AND THE GROWTH OF ALGAE IN LAKE KINNERET 1 , 1975 .

[14]  Friedrich Recknagel,et al.  Automated modelling of a food web in lake Bled using measured data and a library of domain knowledge , 2006 .

[15]  Justinian P. Rosca.  Proceedings of the Workshop on Genetic Programming: From Theory to Real-World Applications , 1995 .

[16]  Peter A. Whigham,et al.  Grammatically-based Genetic Programming , 1995 .

[17]  Yosef Z. Yacobi,et al.  Primary production and phytoplankton in Lake Kinneret: A long‐term record (1972‐1993) , 1995 .

[18]  Saso Dzeroski,et al.  Inducing Process Models from Continuous Data , 2002, ICML.

[19]  Nataša Atanasova.  Preparation and use of the domain expert knowledge for automated modelling of aquatic ecosystems , 2005 .

[20]  David P. Hamilton,et al.  Ten steps applied to development and evaluation of process-based biogeochemical models of estuaries , 2008, Environ. Model. Softw..

[21]  P. Verhulst.  Recherches mathématiques sur la loi d’accroissement de la population , 1845, Nouveaux mémoires de l'Académie royale des sciences et belles-lettres de Bruxelles.

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

[23]  Q. Dortch,et al.  The interaction between ammonium and nitrate uptake in phytoplankton , 1990 .

[24]  Tamar Zohary,et al.  Alternative states in the phytoplankton of Lake Kinneret, Israel (Sea of Galilee) , 2007 .

[25]  Saso Dzeroski,et al.  Declarative Bias in Equation Discovery , 1997, ICML.

[26]  Sašo Džeroski,et al.  Application of automated model discovery from data and expert knowledge to a real-world domain: Lake Glumsø , 2008 .

[27]  W. Rodhe,et al.  Selenium as a micronutrient for the dinoflagellate Peridinium cinctum fa. westii: With 4 figures and 1 table in the text , 1978 .

[28]  Anthony J. Jakeman,et al.  Ten iterative steps in development and evaluation of environmental models , 2006, Environ. Model. Softw..

[29]  A. Kaplan,et al.  Programmed cell death of the dinoflagellate Peridinium gatunense is mediated by CO2 limitation and oxidative stress , 1999, Current Biology.

[30]  Donald Scavia,et al.  An ecological model of Lake Ontario , 1980 .

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

[32]  David P. Hamilton,et al.  A numerical simulation of the role of zooplankton in C, N and P cycling in Lake Kinneret, Israel , 2006 .

[33]  B. Cavari,et al.  ATP in Lake Kinneret: Indicator of microbial biomass or of phosphorus deficiency?1 , 1976 .