A decision support system based on Bayesian modelling for pest management: Application to wireworm risk assessment in maize fields
Protecting crops against pests is a major issue in the current agricultural production system. In particular, assessing the risk to crops can promote integrated pest management (IPM) strategies that encourage natural control mechanisms and advocate the use of pesticides as a last resort. In this stu...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-08-01
|
| Series: | Smart Agricultural Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772375522001265 |
| _version_ | 1827958926522449920 |
|---|---|
| author | Julien Roche Manuel Plantegenest Philippe Larroudé Jean-Baptiste Thibord Le Cointe Ronan Sylvain Poggi |
| author_facet | Julien Roche Manuel Plantegenest Philippe Larroudé Jean-Baptiste Thibord Le Cointe Ronan Sylvain Poggi |
| author_sort | Julien Roche |
| collection | DOAJ |
| description | Protecting crops against pests is a major issue in the current agricultural production system. In particular, assessing the risk to crops can promote integrated pest management (IPM) strategies that encourage natural control mechanisms and advocate the use of pesticides as a last resort. In this study, we focused on wireworms, major soil-dwelling insect pests inflicting severe economic damage on various crops (including maize, potatoes and cereals) across Europe and North America. We have developed an original hierarchical Bayesian model that explicitly accounts for biological knowledge and uncertainty in field observations, rather than relying solely on statistical correlations, to predict the level of wireworm infestation. The model was calibrated and validated using a substantial dataset originating from an agro-environmental survey carried out over three consecutive years (2012–2014) in France, which provides the wireworm abundance in 419 maize fields, together with information on the landscape context, field history, weather conditions, soil characteristics and farming practices associated to each field. Model outcomes show good agreement with current knowledge from literature and field expertise in terms of the effects of variables on wireworm abundance, and provide fairly good predictive capacity. Subsequently, the model was encapsulated as a software (R shiny application) to predict the risk of wireworm infestation in any field of interest, and can be used by farmers or agricultural advisors as a decision support system for the implementation of IPM strategies. The conceptual framework that we implemented can be adapted to a wide range of similar situations involving other crops and pests. |
| first_indexed | 2024-04-09T15:43:37Z |
| format | Article |
| id | doaj.art-ee67a610e72243b0a6fc5a90bd1b6ada |
| institution | Directory Open Access Journal |
| issn | 2772-3755 |
| language | English |
| last_indexed | 2024-04-09T15:43:37Z |
| publishDate | 2023-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Smart Agricultural Technology |
| spelling | doaj.art-ee67a610e72243b0a6fc5a90bd1b6ada2023-04-27T06:08:25ZengElsevierSmart Agricultural Technology2772-37552023-08-014100162A decision support system based on Bayesian modelling for pest management: Application to wireworm risk assessment in maize fieldsJulien Roche0Manuel Plantegenest1Philippe Larroudé2Jean-Baptiste Thibord3Le Cointe Ronan4Sylvain Poggi5IGEPP, INRAE, Institut Agro, Univ Rennes, Le Rheu 35653, FranceIGEPP, INRAE, Institut Agro, Univ Rennes, Le Rheu 35653, France; IGEPP, INRAE, Institut Agro, Univ Rennes, Rennes 35000, FranceArvalis Institut du Végétal, Agrosite 21 chemin de Pau, Montardon 64121, FranceArvalis Institut du Végétal, Agrosite 21 chemin de Pau, Montardon 64121, FranceIGEPP, INRAE, Institut Agro, Univ Rennes, Le Rheu 35653, FranceIGEPP, INRAE, Institut Agro, Univ Rennes, Le Rheu 35653, France; Corresponding author.Protecting crops against pests is a major issue in the current agricultural production system. In particular, assessing the risk to crops can promote integrated pest management (IPM) strategies that encourage natural control mechanisms and advocate the use of pesticides as a last resort. In this study, we focused on wireworms, major soil-dwelling insect pests inflicting severe economic damage on various crops (including maize, potatoes and cereals) across Europe and North America. We have developed an original hierarchical Bayesian model that explicitly accounts for biological knowledge and uncertainty in field observations, rather than relying solely on statistical correlations, to predict the level of wireworm infestation. The model was calibrated and validated using a substantial dataset originating from an agro-environmental survey carried out over three consecutive years (2012–2014) in France, which provides the wireworm abundance in 419 maize fields, together with information on the landscape context, field history, weather conditions, soil characteristics and farming practices associated to each field. Model outcomes show good agreement with current knowledge from literature and field expertise in terms of the effects of variables on wireworm abundance, and provide fairly good predictive capacity. Subsequently, the model was encapsulated as a software (R shiny application) to predict the risk of wireworm infestation in any field of interest, and can be used by farmers or agricultural advisors as a decision support system for the implementation of IPM strategies. The conceptual framework that we implemented can be adapted to a wide range of similar situations involving other crops and pests.http://www.sciencedirect.com/science/article/pii/S2772375522001265Pest risk assessmentIntegrated pest management (IPM)Hierarchical Bayesian modelDecision support systemWireworm controlCrop protection |
| spellingShingle | Julien Roche Manuel Plantegenest Philippe Larroudé Jean-Baptiste Thibord Le Cointe Ronan Sylvain Poggi A decision support system based on Bayesian modelling for pest management: Application to wireworm risk assessment in maize fields Smart Agricultural Technology Pest risk assessment Integrated pest management (IPM) Hierarchical Bayesian model Decision support system Wireworm control Crop protection |
| title | A decision support system based on Bayesian modelling for pest management: Application to wireworm risk assessment in maize fields |
| title_full | A decision support system based on Bayesian modelling for pest management: Application to wireworm risk assessment in maize fields |
| title_fullStr | A decision support system based on Bayesian modelling for pest management: Application to wireworm risk assessment in maize fields |
| title_full_unstemmed | A decision support system based on Bayesian modelling for pest management: Application to wireworm risk assessment in maize fields |
| title_short | A decision support system based on Bayesian modelling for pest management: Application to wireworm risk assessment in maize fields |
| title_sort | decision support system based on bayesian modelling for pest management application to wireworm risk assessment in maize fields |
| topic | Pest risk assessment Integrated pest management (IPM) Hierarchical Bayesian model Decision support system Wireworm control Crop protection |
| url | http://www.sciencedirect.com/science/article/pii/S2772375522001265 |
| work_keys_str_mv | AT julienroche adecisionsupportsystembasedonbayesianmodellingforpestmanagementapplicationtowirewormriskassessmentinmaizefields AT manuelplantegenest adecisionsupportsystembasedonbayesianmodellingforpestmanagementapplicationtowirewormriskassessmentinmaizefields AT philippelarroude adecisionsupportsystembasedonbayesianmodellingforpestmanagementapplicationtowirewormriskassessmentinmaizefields AT jeanbaptistethibord adecisionsupportsystembasedonbayesianmodellingforpestmanagementapplicationtowirewormriskassessmentinmaizefields AT lecointeronan adecisionsupportsystembasedonbayesianmodellingforpestmanagementapplicationtowirewormriskassessmentinmaizefields AT sylvainpoggi adecisionsupportsystembasedonbayesianmodellingforpestmanagementapplicationtowirewormriskassessmentinmaizefields AT julienroche decisionsupportsystembasedonbayesianmodellingforpestmanagementapplicationtowirewormriskassessmentinmaizefields AT manuelplantegenest decisionsupportsystembasedonbayesianmodellingforpestmanagementapplicationtowirewormriskassessmentinmaizefields AT philippelarroude decisionsupportsystembasedonbayesianmodellingforpestmanagementapplicationtowirewormriskassessmentinmaizefields AT jeanbaptistethibord decisionsupportsystembasedonbayesianmodellingforpestmanagementapplicationtowirewormriskassessmentinmaizefields AT lecointeronan decisionsupportsystembasedonbayesianmodellingforpestmanagementapplicationtowirewormriskassessmentinmaizefields AT sylvainpoggi decisionsupportsystembasedonbayesianmodellingforpestmanagementapplicationtowirewormriskassessmentinmaizefields |