Spatial and climatic factors associated with the geographical distribution of citrus black spot disease in South Africa. A Bayesian latent Gaussian model approach
Citrus black spot (CBS), caused by Phyllosticta citricarpa, is the main fungal disease affecting this crop and quarantine measures have been implemented. The role of climate as a limiting factor for the establishment and spread of CBS to new areas has been debated, but previous studies did not addre...
| Autores principales: | , , , |
|---|---|
| Formato: | article |
| Lenguaje: | Inglés |
| Publicado: |
2019
|
| Acceso en línea: | http://hdl.handle.net/20.500.11939/6230 |
| _version_ | 1855032389980389376 |
|---|---|
| author | Martínez-Minaya, Joaquín Conesa, David López-Quílez, Antonio Vicent, Antonio |
| author_browse | Conesa, David López-Quílez, Antonio Martínez-Minaya, Joaquín Vicent, Antonio |
| author_facet | Martínez-Minaya, Joaquín Conesa, David López-Quílez, Antonio Vicent, Antonio |
| author_sort | Martínez-Minaya, Joaquín |
| collection | ReDivia |
| description | Citrus black spot (CBS), caused by Phyllosticta citricarpa, is the main fungal disease affecting this crop and quarantine measures have been implemented. The role of climate as a limiting factor for the establishment and spread of CBS to new areas has been debated, but previous studies did not address the effects of spatial factors in the geographic distribution of the disease. The effects of climatic and spatial factors were studied using South Africa as a case study, due to its diversity of climates within citrus-growing regions. Georeferenced datasets of CBS presence/absence in citrus areas were assembled for two stages of the epidemic: 1950 and 2014. Climatic variables were obtained from the WorldClim database. Moran's I and Geary's C analyses indicated that CBS distribution data presented significant spatial autocorrelation, particularly in 2014. Collinearity was detected among climatic variables. Spatial logistic regressions, particular case of latent Gaussian models, were fitted to CBS presence/absence in 1950 or 2014 with the Integrated Nested Laplace Approximation methodology. Principal components (PCs) or pre-selection of climatic variables based on their correlation coefficients were used to cope with collinearity. Spatial effects were incorporated with a geostatistical term. In general, the models indicated a positive relationship between CBS presence and climatic variables or PCs associated with warm temperatures and high precipitation. Nevertheless, in 1950, models that also included a spatial effect outperformed those with climatic variables only. Problems of model convergence were detected in 2014 due to the strong spatial structure of CBS distribution data. The consequences of ignoring spatial dependence to estimate the potential geographical range of CBS are discussed. |
| format | article |
| id | ReDivia6230 |
| institution | Instituto Valenciano de Investigaciones Agrarias (IVIA) |
| language | Inglés |
| publishDate | 2019 |
| publishDateRange | 2019 |
| publishDateSort | 2019 |
| record_format | dspace |
| spelling | ReDivia62302025-04-25T14:46:35Z Spatial and climatic factors associated with the geographical distribution of citrus black spot disease in South Africa. A Bayesian latent Gaussian model approach Martínez-Minaya, Joaquín Conesa, David López-Quílez, Antonio Vicent, Antonio Citrus black spot (CBS), caused by Phyllosticta citricarpa, is the main fungal disease affecting this crop and quarantine measures have been implemented. The role of climate as a limiting factor for the establishment and spread of CBS to new areas has been debated, but previous studies did not address the effects of spatial factors in the geographic distribution of the disease. The effects of climatic and spatial factors were studied using South Africa as a case study, due to its diversity of climates within citrus-growing regions. Georeferenced datasets of CBS presence/absence in citrus areas were assembled for two stages of the epidemic: 1950 and 2014. Climatic variables were obtained from the WorldClim database. Moran's I and Geary's C analyses indicated that CBS distribution data presented significant spatial autocorrelation, particularly in 2014. Collinearity was detected among climatic variables. Spatial logistic regressions, particular case of latent Gaussian models, were fitted to CBS presence/absence in 1950 or 2014 with the Integrated Nested Laplace Approximation methodology. Principal components (PCs) or pre-selection of climatic variables based on their correlation coefficients were used to cope with collinearity. Spatial effects were incorporated with a geostatistical term. In general, the models indicated a positive relationship between CBS presence and climatic variables or PCs associated with warm temperatures and high precipitation. Nevertheless, in 1950, models that also included a spatial effect outperformed those with climatic variables only. Problems of model convergence were detected in 2014 due to the strong spatial structure of CBS distribution data. The consequences of ignoring spatial dependence to estimate the potential geographical range of CBS are discussed. 2019-05-15T10:37:42Z 2019-05-15T10:37:42Z 2018 article Martínez-Minaya, J.; Conesa, D.; López-Quilez, A.; Vicent, A. (2018). Spatial and climatic factors associated with the geographical distribution of citrus black spot disease in south africa. A bayesian latent gaussian model approach. European Journal of Plant Pathology, 151(4), 991-1007. http://hdl.handle.net/20.500.11939/6230 10.1007/s10658-018-1435-6 en Atribución-NoComercial-SinDerivadas 3.0 España http://creativecommons.org/licenses/by-nc-nd/3.0/es/ electronico |
| spellingShingle | Martínez-Minaya, Joaquín Conesa, David López-Quílez, Antonio Vicent, Antonio Spatial and climatic factors associated with the geographical distribution of citrus black spot disease in South Africa. A Bayesian latent Gaussian model approach |
| title | Spatial and climatic factors associated with the geographical distribution of citrus black spot disease in South Africa. A Bayesian latent Gaussian model approach |
| title_full | Spatial and climatic factors associated with the geographical distribution of citrus black spot disease in South Africa. A Bayesian latent Gaussian model approach |
| title_fullStr | Spatial and climatic factors associated with the geographical distribution of citrus black spot disease in South Africa. A Bayesian latent Gaussian model approach |
| title_full_unstemmed | Spatial and climatic factors associated with the geographical distribution of citrus black spot disease in South Africa. A Bayesian latent Gaussian model approach |
| title_short | Spatial and climatic factors associated with the geographical distribution of citrus black spot disease in South Africa. A Bayesian latent Gaussian model approach |
| title_sort | spatial and climatic factors associated with the geographical distribution of citrus black spot disease in south africa a bayesian latent gaussian model approach |
| url | http://hdl.handle.net/20.500.11939/6230 |
| work_keys_str_mv | AT martinezminayajoaquin spatialandclimaticfactorsassociatedwiththegeographicaldistributionofcitrusblackspotdiseaseinsouthafricaabayesianlatentgaussianmodelapproach AT conesadavid spatialandclimaticfactorsassociatedwiththegeographicaldistributionofcitrusblackspotdiseaseinsouthafricaabayesianlatentgaussianmodelapproach AT lopezquilezantonio spatialandclimaticfactorsassociatedwiththegeographicaldistributionofcitrusblackspotdiseaseinsouthafricaabayesianlatentgaussianmodelapproach AT vicentantonio spatialandclimaticfactorsassociatedwiththegeographicaldistributionofcitrusblackspotdiseaseinsouthafricaabayesianlatentgaussianmodelapproach |