Simplifying a complex computer model: Sensitivity analysis and metamodelling of an 3D individual-based crop-weed canopy model

Abstract Complex biological models such as mechanistic research models often need to extend their current use to a broader audience. Simplification and faster simulations would increase their use. Here, a step-by-step methodology was developed and applied to partially metamodel, hence accelerate, the mechanistic model FlorSys . This is a process-based, multiannual and multispecies model ("virtual field") which simulates crop growth and weed dynamics and allows users to assess cropping systems for crop production and biodiversity. The model is relatively slow, which makes it difficult to test numerous and diverse cropping systems needed to identify those reconciling crop production and biodiversity. Here, we (1) identified the slowest submodel of FlorSys , i.e. the 3D voxelized light interception submodel, (2) identified and applied a relevant methodology to metamodel this submodel in the simplest situation, i.e. we predicted light interception and absorption directly at the scale of the plant instead of the voxel for a single plant in a field, and (3) extrapolated the method to more complex situations, i.e. a plant in diverse and heterogeneous crop:weed canopies, (4) replaced the original process-based FlorSys submodel by the metamodels, which required additional equations and decision rules, (5) evaluated the metamodelled FlorSys with independent field observations, showing an adequate prediction quality combined with an increased speed at fine-grained scale since the metamodelled version was 28 times faster than the process-based version. For steps 2 and 3, we used the global sensitivity method based on a truncated Legendre polynomial chaos expansion (PCE) whose coefficients were estimated by Partial Least Squares (PLS) regression to simultaneously (i) rank inputs with respect to their polynomial and total effects on outputs via the so-called PCE-PLS sensitivity indices, and (ii) provide metamodels predicting light interception and absorption at the plant level. These metamodels were then shortened into parsimonious metamodels via a LASSO-PLS method. The study showed that there was a trade-off between speed gain due to the metamodelled 3D light submodel and the speed loss due to the additional functions for neighbourhood effects. The metamodelled version is best used for testing complex systems where plant location must be modelled precisley (e.g., precision agriculture, intercropping with precision sowing) whereas the voxelized version with a large voxel size is better for simpler cropping systems. The present step-by-step process may be helpful for investigating and speeding up other complex simulation models with interacting objects/agents. It notably uses a hybrid approach, using a process-based (albeit simplified) approach for the most sensitive plant stage (newly emerged tiny plants) and separate sampling plans and metamodels to ensure that the more sensitive stages/components are adequately covered (small plants).

[1]  John R. Williams,et al.  A general, process-oriented model for two competing plant species , 1992 .

[2]  Stéphanie Mahévas,et al.  Analyse de sensibilité et exploration de modèles , 2013 .

[3]  J. Villerd,et al.  The FLORSYS crop-weed canopy model, a tool to investigate and promote agroecological weed management , 2021 .

[4]  M. Liebman,et al.  9 – Many Little Hammers: Ecological Management of Crop-Weed Interactions , 1997 .

[5]  L. K. Ward,et al.  The role of weeds in supporting biological diversity within crop fields , 2003 .

[6]  P. Jamieson,et al.  Simplifying Sirius: sensitivity analysis and development of a meta-model for wheat yield prediction , 2001 .

[7]  Michael Renton,et al.  How much detail and accuracy is required in plant growth sub-models to address questions about optimal management strategies in agricultural systems? , 2011, AoB PLANTS.

[8]  S. Wold,et al.  PLS-regression: a basic tool of chemometrics , 2001 .

[9]  Michael Renton Shifting focus from the population to the individual as a way forward in understanding, predicting and managing the complexities of evolution of resistance to pesticides. , 2013, Pest management science.

[10]  M. Monsi Uber den Lichtfaktor in den Pflanzengesellschaften und seine Bedeutung fur die Stoffproduktion , 1953 .

[11]  M. Marek,et al.  Effect of the relative time of emergence on the growth allometry of Galium aparine in competition with Triticum aestivum , 2014 .

[12]  Nathalie Villa-Vialaneix,et al.  A comparison of eight metamodeling techniques for the simulation of N2O fluxes and N leaching from corn crops , 2012, Environ. Model. Softw..

[13]  N. Boatman,et al.  Ecological impacts of early 21st century agricultural change in Europe--a review. , 2009, Journal of environmental management.

[14]  Nathalie Colbach,et al.  Uncertainty analysis and evaluation of a complex, multi-specific weed dynamics model with diverse and incomplete data sets , 2016, Environ. Model. Softw..

[15]  David M Umbach,et al.  Pesticide use and fatal injury among farmers in the Agricultural Health Study , 2013, International Archives of Occupational and Environmental Health.

[16]  Regis Chikowo,et al.  Integrated Weed Management systems allow reduced reliance on herbicides and long-term weed control , 2009 .

[17]  E. Justes,et al.  Current knowledge and future research opportunities for modeling annual crop mixtures. A review , 2019, Agronomy for Sustainable Development.

[18]  Gail G. Wilkerson,et al.  Weed management decision models: pitfalls, perceptions, and possibilities of the economic threshold approach , 2002 .

[19]  R. Plackett,et al.  THE DESIGN OF OPTIMUM MULTIFACTORIAL EXPERIMENTS , 1946 .

[20]  N. Munier-Jolain,et al.  A 3D model for light interception in heterogeneous crop:weed canopies: Model structure and evaluation , 2013 .

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

[22]  N. Munier-Jolain,et al.  Assessing innovative sowing patterns for integrated weed management with a 3D crop:weed competition model , 2014 .

[23]  N. Colbach,et al.  Modeling the dynamics and emergence of a multispecies weed seed bank with species traits , 2012 .

[24]  Eric Justes,et al.  Designing crop management systems by simulation , 2007 .

[25]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[26]  Santiago L. Poggio,et al.  Simulation Models on the Ecology and Management of Arable Weeds: Structure, Quantitative Insights, and Applications , 2020, Agronomy.

[27]  B. Graf,et al.  A simulation model for the dynamics of rice growth and development: Part II—The competition with weeds for nitrogen and light , 1990 .

[28]  N. Colbach Modelling cropping system effects on crop pest dynamics : How to compromise between process analysis and decision aid , 2010 .

[29]  Philip A. Stephens,et al.  Predictive models of weed population dynamics , 2009 .

[30]  A. Saltelli,et al.  Making best use of model evaluations to compute sensitivity indices , 2002 .

[31]  P. Debaeke,et al.  Integrating crop management and crop rotation effects into models of weed population dynamics: a review , 1998, Weed Science.

[32]  Sanjay B. Joshi,et al.  Metamodeling: Radial basis functions, versus polynomials , 2002, Eur. J. Oper. Res..

[33]  L. R. Benjamin,et al.  An evaluation of four crop : weed competition models using a common data set , 2003 .

[34]  Niels Holst,et al.  Field weed population dynamics : a review of model approaches and applications , 2007 .

[35]  Martin J. Kropff,et al.  An eco-physiological model for interspecific competition, applied to the influence of Chenopodium album L. on sugar beet. I : Model description and parameterization , 1992 .

[36]  W. de Groot,et al.  An eco-physiological model for interspecific competition, applied to the influence of Chenopodium album L. on sugar beet. II. Model evaluation , 1992 .

[37]  Nicolas Munier-Jolain,et al.  Weeds in agricultural landscapes. A review , 2011, Agronomy for Sustainable Development.

[38]  J. Villerd,et al.  A methodology for multi-objective cropping system design based on simulations. Application to weed management , 2017 .

[39]  Wei-Yin Loh,et al.  Classification and regression trees , 2011, WIREs Data Mining Knowl. Discov..

[40]  Chien Wang,et al.  Metamodeling of Droplet Activation for Global Climate Models , 2016 .

[41]  B. Bryan,et al.  Meta-modeling soil organic carbon sequestration potential and its application at regional scale. , 2013, Ecological applications : a publication of the Ecological Society of America.

[42]  Wan,et al.  Evaluation of neural network models with generalized sensitivity analysis , 2000, Analytical chemistry.

[43]  Jack P. C. Kleijnen,et al.  A methodology for fitting and validating metamodels in simulation , 2000, Eur. J. Oper. Res..

[44]  Jean-Pierre Gauchi,et al.  Selecting both latent and explanatory variables in the PLS1 regression model , 2003 .

[45]  Modelling crop:weed interactions in wheat with ALMANAC , 1997 .

[46]  Richard J. Beckman,et al.  A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output From a Computer Code , 2000, Technometrics.

[47]  Nathalie Colbach,et al.  Co-development of a decision support system for integrated weed management: Contribution from future users , 2020 .

[48]  M. Monsi,et al.  On the factor light in plant communities and its importance for matter production. 1953. , 2004, Annals of botany.

[49]  L. Hossard,et al.  Meta-modelling of the impacts of regional cropping system scenarios for phoma stem canker control , 2015 .

[50]  E. Oerke Crop losses to pests , 2005, The Journal of Agricultural Science.

[51]  James W. Jones,et al.  SOYWEED: A Simulation Model of Soybean and Common Cocklebur Growth and Competition , 1990 .

[52]  J. Netland,et al.  Towards machine vision based site-specific weed management in cereals , 2012 .

[53]  R. Iman,et al.  A distribution-free approach to inducing rank correlation among input variables , 1982 .

[54]  Bruno Sudret,et al.  Global sensitivity analysis using polynomial chaos expansions , 2008, Reliab. Eng. Syst. Saf..

[55]  Nathalie Colbach,et al.  Metamodeling and global sensitivity analysis for computer models with correlated inputs: A practical approach tested with a 3D light interception computer model , 2017, Environ. Model. Softw..

[56]  Development of a fast, urban chemistry metamodel for inclusion in global models , 2011 .

[57]  Guillaume Marie,et al.  Extending the use of ecological models without sacrificing details: a generic and parsimonious meta‐modelling approach , 2014 .

[58]  R. Tibshirani Regression Shrinkage and Selection via the Lasso , 1996 .