Victor Dufleit
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Victor Dufleit has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Eric Etter
ASTRE, Univ Montpellier, CIRAD, INRAE, Montpellier, France
5 shared publications - Laure Guerrini
ASTRE, CIRAD, INRAe, Université Montpellier, Montpellier, France
4 shared publications Cozier Joanna
2 shared publications- Damien F. Meyer
CIRAD, UMR ASTRE, F-1970 Petit-Bourg – Guadeloupe, France
2 shared publications Daniele Da Re
Center Agriculture Food Environment, University of Trento, San Michele all' Adige, Italy
2 shared publicationsMarius Gilbert
Universite Libre de Bruxelles
2 shared publications
Latest Publications
- 10 June 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Epidémiologie de la cowdriose : modélisations des risques et perspectives Caraïbéennes
La cowdriose est une maladie infectieuse des ruminants causée par la bactérie Ehrlichia ruminantium (famille des Anaplasmataceae), transmise par les tiques du genre Amblyomma. Principalement décrite en Afrique, elle est également présente dans la Caraïbe, où elle aurait été introduite durant la période du commerce triangulaire, par l’import de bovidés d’Afrique de l’ouest, infestés par la tique Amblyomma variegatum, notamment en Guadeloupe. L’archipel de Guadeloupe, tout comme l’île d’Antigua, sont les seules zones où la présence de cowdriose a pu être confirmée dans la Caraïbe. Au cours de la seconde moitié du 20ᵉ siècle, l’extension progressive d’A. variegatum d’île en île a été observée, probablement facilitée par les déplacements de hérons garde-bœufs, colonisateurs récents des Amériques.Cette expansion dans la Caraïbe laisse aussi craindre une potentielle introduction du pathogène E. ruminantium dans des territoires actuellement indemnes, notamment les États-Unis. Divers scénarios d’introduction via les hérons migrateurs ou d’autres voies apparaissent plausibles. Les conséquences d’une telle introduction pourraient être majeures, compte tenu de la forte sensibilité des ruminants américains à la maladie et la capacité de certaines tiques largement distribuées sur le continent américain à pouvoir devenir de potentiels réservoirs du pathogène de la cowdriose.Cette thèse visait à évaluer le risque d’introduction d’Ehrlichia ruminantium aux États-Unis depuis la Caraïbe, ainsi qu’à actualiser les données épidémiologiques de la cowdriose dans la région. En l’absence de données disponibles en dehors de l’archipel de Guadeloupe, les efforts se sont donc concentrés sur cette île, dans une logique de développement d’une méthodologie territorialisée, transférable à d’autres îles caribéennes. L’analyse du risque d’introduction sur le territoire états-unien a été conduite de manière semi-quantitative, en s’appuyant notamment sur une élicitation d’experts pour pallier l’absence de données.Une enquête épidémiologique de terrain a été menée auprès des éleveurs bovins guadeloupéens. Elle a permis de recueillir des informations sur les pratiques d’élevage, de réaliser des inspections animales, ainsi que des prélèvements sanguins pour caractériser les infestations par Amblyomma variegatum et Rhipicephalus microplus, et d’évaluer le statut immunologique des ruminants vis-à-vis de la cowdriose. Des prélèvements échantillonnés de manière non probabiliste ont également été réalisés sur des petits ruminants afin de compléter notre échantillonnage, notamment sur les îles des Saintes et de la Désirade. Les résultats de l’enquête ont permis une mise à jour des données de prévalence, et mis en évidence, pour la première fois, la circulation d’E. ruminantium dans l’archipel des Saintes, jusque-là considéré comme indemne. Des analyses de facteurs de risque ont été conduites à partir des données collectées, tant pour la présence de cowdriose que pour les infestations par les deux espèces de tiques, révélant des différences notables entre les îles de l’archipel.L’un des objectifs de la thèse était également de produire un indicateur spatial du risque de cowdriose. Une première étape a consisté à générer une carte de densité bovine à haute résolution, en adaptant la méthode de désagrégation des recensements développée par la FAO (GLW) aux données locales. Les données d’infestation ont ensuite permis de modéliser les niches écologiques des deux espèces de tiques caractérisant leur distribution sur le territoire. Les cartes d’hôtes et de vecteurs ont été intégrées dans une analyse de décision multicritère (MCDA), conduisant à l’élaboration d’un indicateur spatial composite du risque de cowdriose en Guadeloupe.Enfin, une analyse semi-quantitative du risque d’introduction de la cowdriose vers les États-Unis a été conduite. Différents scénarios d’introduction ont été formulés à partir de la littérature et par la mise en place d’une élicitation d’expert.
- 10 June 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Epidémiologie de la cowdriose : modélisations des risques et perspectives Caraïbéennes
La cowdriose est une maladie infectieuse des ruminants causée par la bactérie Ehrlichia ruminantium (famille des Anaplasmataceae), transmise par les tiques du genre Amblyomma. Principalement décrite en Afrique, elle est également présente dans la Caraïbe, où elle aurait été introduite durant la période du commerce triangulaire, par l’import de bovidés d’Afrique de l’ouest, infestés par la tique Amblyomma variegatum, notamment en Guadeloupe. L’archipel de Guadeloupe, tout comme l’île d’Antigua, sont les seules zones où la présence de cowdriose a pu être confirmée dans la Caraïbe. Au cours de la seconde moitié du 20ᵉ siècle, l’extension progressive d’A. variegatum d’île en île a été observée, probablement facilitée par les déplacements de hérons garde-bœufs, colonisateurs récents des Amériques.Cette expansion dans la Caraïbe laisse aussi craindre une potentielle introduction du pathogène E. ruminantium dans des territoires actuellement indemnes, notamment les États-Unis. Divers scénarios d’introduction via les hérons migrateurs ou d’autres voies apparaissent plausibles. Les conséquences d’une telle introduction pourraient être majeures, compte tenu de la forte sensibilité des ruminants américains à la maladie et la capacité de certaines tiques largement distribuées sur le continent américain à pouvoir devenir de potentiels réservoirs du pathogène de la cowdriose.Cette thèse visait à évaluer le risque d’introduction d’Ehrlichia ruminantium aux États-Unis depuis la Caraïbe, ainsi qu’à actualiser les données épidémiologiques de la cowdriose dans la région. En l’absence de données disponibles en dehors de l’archipel de Guadeloupe, les efforts se sont donc concentrés sur cette île, dans une logique de développement d’une méthodologie territorialisée, transférable à d’autres îles caribéennes. L’analyse du risque d’introduction sur le territoire états-unien a été conduite de manière semi-quantitative, en s’appuyant notamment sur une élicitation d’experts pour pallier l’absence de données.Une enquête épidémiologique de terrain a été menée auprès des éleveurs bovins guadeloupéens. Elle a permis de recueillir des informations sur les pratiques d’élevage, de réaliser des inspections animales, ainsi que des prélèvements sanguins pour caractériser les infestations par Amblyomma variegatum et Rhipicephalus microplus, et d’évaluer le statut immunologique des ruminants vis-à-vis de la cowdriose. Des prélèvements échantillonnés de manière non probabiliste ont également été réalisés sur des petits ruminants afin de compléter notre échantillonnage, notamment sur les îles des Saintes et de la Désirade. Les résultats de l’enquête ont permis une mise à jour des données de prévalence, et mis en évidence, pour la première fois, la circulation d’E. ruminantium dans l’archipel des Saintes, jusque-là considéré comme indemne. Des analyses de facteurs de risque ont été conduites à partir des données collectées, tant pour la présence de cowdriose que pour les infestations par les deux espèces de tiques, révélant des différences notables entre les îles de l’archipel.L’un des objectifs de la thèse était également de produire un indicateur spatial du risque de cowdriose. Une première étape a consisté à générer une carte de densité bovine à haute résolution, en adaptant la méthode de désagrégation des recensements développée par la FAO (GLW) aux données locales. Les données d’infestation ont ensuite permis de modéliser les niches écologiques des deux espèces de tiques caractérisant leur distribution sur le territoire. Les cartes d’hôtes et de vecteurs ont été intégrées dans une analyse de décision multicritère (MCDA), conduisant à l’élaboration d’un indicateur spatial composite du risque de cowdriose en Guadeloupe.Enfin, une analyse semi-quantitative du risque d’introduction de la cowdriose vers les États-Unis a été conduite. Différents scénarios d’introduction ont été formulés à partir de la littérature et par la mise en place d’une élicitation d’expert.
- Open Access
- 21 January 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (1) , e0324695
Downscaling local distribution of cattle over Guadeloupe archipelago: An adapted method for disaggregating census data
Gridded livestock distribution datasets have been produced for several years and are used in various fields, including epidemiology, livestock impact assessment, and territory management. Those datasets are based on census conducted at national/sub-national scale which are then downscaled using machine learning algorithm and relevant spatial-explicit environmental predictors. The most known dataset of livestock disaggregated observations is the Gridded Livestock of the World (GLW), which produces global maps of livestock density at 10 km spatial resolution for several livestock species. Though this spatial resolution can be appropriate to describe livestock distribution at the global scale, it inherently leads to a coarse representation of breeding species density for smaller territories such as the Caribbean Islands. In this study, we propose an adaptation of the GLW methodology that accounts for the spatial autocorrelation in observed cattle distribution, thereby better capturing the specific characteristics of geographically limited areas such as the Guadeloupean archipelago. Cattle census data were collected for the 32 municipalities of the archipelago and associated to environmental predictors derived from remote sensing and land cover datasets. Together with the Random Forest (RF) algorithm used in the standard GLW methodology, we tested the performance of a Geographical Random Forest (GRF), a novel methodology allowing for taking into account the spatial autocorrelation of the response variable. The GRF algorithm demonstrated significantly better performance compared to the RF algorithm, albeit with longer processing times, and allowed us producing cattle distribution maps for the entire Guadeloupe archipelago at a spatial resolution of 225 m using both algorithms. The approach developed holds potential for application to other small territories, including other islands in the Caribbean.
- 21 January 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (1) , e0324695
Open AccessDownscaling local distribution of cattle over Guadeloupe archipelago: An adapted method for disaggregating census data
Gridded livestock distribution datasets have been produced for several years and are used in various fields, including epidemiology, livestock impact assessment, and territory management. Those datasets are based on census conducted at national/sub-national scale which are then downscaled using machine learning algorithm and relevant spatial-explicit environmental predictors. The most known dataset of livestock disaggregated observations is the Gridded Livestock of the World (GLW), which produces global maps of livestock density at 10 km spatial resolution for several livestock species. Though this spatial resolution can be appropriate to describe livestock distribution at the global scale, it inherently leads to a coarse representation of breeding species density for smaller territories such as the Caribbean Islands. In this study, we propose an adaptation of the GLW methodology that accounts for the spatial autocorrelation in observed cattle distribution, thereby better capturing the specific characteristics of geographically limited areas such as the Guadeloupean archipelago. Cattle census data were collected for the 32 municipalities of the archipelago and associated to environmental predictors derived from remote sensing and land cover datasets. Together with the Random Forest (RF) algorithm used in the standard GLW methodology, we tested the performance of a Geographical Random Forest (GRF), a novel methodology allowing for taking into account the spatial autocorrelation of the response variable. The GRF algorithm demonstrated significantly better performance compared to the RF algorithm, albeit with longer processing times, and allowed us producing cattle distribution maps for the entire Guadeloupe archipelago at a spatial resolution of 225 m using both algorithms. The approach developed holds potential for application to other small territories, including other islands in the Caribbean.
- Open Access
- 20 November 2025
- Published by Elsevier in Ticks and Tick-borne Diseases
- Vol. 16 (6) , 102569
Prevalence of heartwater in Guadeloupe (2024): stable endemicity and evidence of spread to Les Saintes
- 20 November 2025
- Published by Elsevier in Ticks and Tick-borne Diseases
- Vol. 16 (6) , 102569
Open AccessPrevalence of heartwater in Guadeloupe (2024): stable endemicity and evidence of spread to Les Saintes
- Open Access
Risk factors associated with tick infestation and Ehrlichia ruminantium amongst cattle in Guadeloupe, French West Indies
Ehrlichia ruminantium infection and infestations by two tick species, Amblyomma variegatum and Rhipicephalus microplus , are a major threat to the livestock sector in Guadeloupe. This study examined specific tick infestations for 261 cattle from 73 farms on the archipelago’s three largest islands, Basse-Terre, Grande-Terre and Marie-Galante. A questionnaire was implemented to gather information on farming practices and livestock characteristics in order to identify potential risk factors associated with the presence of antibodies against E. ruminantium and the abundance of ticks in livestock. A total of 81% of the inspected cattle were found to be infested with ticks, with a prevalence of 69% for A. variegatum and 47% for R. microplus . At farm level, 96% of farms had infested animals, 84% of which were infested with A. variegatum and 73% with R. microplus . Risk factors were first explored using multiple correspondence analysis (MCA) followed by a generalized linear model at the farm level, and finally a generalized linear mixed model at cattle level. MCA revealed that non-traditional breeding system, with free-grazing animals, were associated with serological evidence of E. ruminantium circulation and also higher tick infestations. The utilisation of green fodder had a significant impact on the prevalence of seropositive animals and tick abundance. Differences in infestation were observed between the three largest islands of Guadeloupe and have been discussed in this article. The frequency of acaricide bathing and its interaction with acaricide alternation appeared to be significant factors in the mitigation of tick infestations in cattle. However, this study highlights the potential emergence of acaricide resistance within tick populations in Guadeloupe.
Open AccessRisk factors associated with tick infestation and Ehrlichia ruminantium amongst cattle in Guadeloupe, French West Indies
Ehrlichia ruminantium infection and infestations by two tick species, Amblyomma variegatum and Rhipicephalus microplus , are a major threat to the livestock sector in Guadeloupe. This study examined specific tick infestations for 261 cattle from 73 farms on the archipelago’s three largest islands, Basse-Terre, Grande-Terre and Marie-Galante. A questionnaire was implemented to gather information on farming practices and livestock characteristics in order to identify potential risk factors associated with the presence of antibodies against E. ruminantium and the abundance of ticks in livestock. A total of 81% of the inspected cattle were found to be infested with ticks, with a prevalence of 69% for A. variegatum and 47% for R. microplus . At farm level, 96% of farms had infested animals, 84% of which were infested with A. variegatum and 73% with R. microplus . Risk factors were first explored using multiple correspondence analysis (MCA) followed by a generalized linear model at the farm level, and finally a generalized linear mixed model at cattle level. MCA revealed that non-traditional breeding system, with free-grazing animals, were associated with serological evidence of E. ruminantium circulation and also higher tick infestations. The utilisation of green fodder had a significant impact on the prevalence of seropositive animals and tick abundance. Differences in infestation were observed between the three largest islands of Guadeloupe and have been discussed in this article. The frequency of acaricide bathing and its interaction with acaricide alternation appeared to be significant factors in the mitigation of tick infestations in cattle. However, this study highlights the potential emergence of acaricide resistance within tick populations in Guadeloupe.
- Open Access
Prevalence of heartwater in Guadeloupe (2024): stable endemicity and evidence of spread to Les Saintes
Heartwater is a tick-borne disease caused by Ehrlichia ruminantium and remains endemic in the Guadeloupe archipelago. While previous data from 1989 suggested a seroprevalence of 22% in cattle, no updated figures were available for livestock, and certain islands such as Les Saintes were considered disease-free. This study aimed to update the sero-epidemiological status of heartwater in cattle and goats across Guadeloupe in 2024. Blood samples were collected from 261 cattle and 135 goats from all islands. Serological testing using the MAP1B ELISA was performed, and true seroprevalence was estimated using a simplified stochastic model that accounts for test sensitivity, specificity, and sample uncertainty. The mean cattle seroprevalence was estimated at 28% (95% CI [22–34%]), with no significant differences between Basse-Terre, Grande-Terre, and Marie-Galante. Importantly, seropositive goats were identified on all major islands, including Les Saintes. Our findings suggest that heartwater remains endemically stable in Guadeloupe since the 1980s and reveal for the first time its serological presence in Les Saintes. These results emphasize the importance of strengthening regional surveillance systems, including veterinary reporting, serological monitoring, and coordinated efforts to prevent further spread across the Caribbean.
Open AccessPrevalence of heartwater in Guadeloupe (2024): stable endemicity and evidence of spread to Les Saintes
Heartwater is a tick-borne disease caused by Ehrlichia ruminantium and remains endemic in the Guadeloupe archipelago. While previous data from 1989 suggested a seroprevalence of 22% in cattle, no updated figures were available for livestock, and certain islands such as Les Saintes were considered disease-free. This study aimed to update the sero-epidemiological status of heartwater in cattle and goats across Guadeloupe in 2024. Blood samples were collected from 261 cattle and 135 goats from all islands. Serological testing using the MAP1B ELISA was performed, and true seroprevalence was estimated using a simplified stochastic model that accounts for test sensitivity, specificity, and sample uncertainty. The mean cattle seroprevalence was estimated at 28% (95% CI [22–34%]), with no significant differences between Basse-Terre, Grande-Terre, and Marie-Galante. Importantly, seropositive goats were identified on all major islands, including Les Saintes. Our findings suggest that heartwater remains endemically stable in Guadeloupe since the 1980s and reveal for the first time its serological presence in Les Saintes. These results emphasize the importance of strengthening regional surveillance systems, including veterinary reporting, serological monitoring, and coordinated efforts to prevent further spread across the Caribbean.
- Open Access
Downscaling local distribution of cattle over Guadeloupe archipelago: an adapted method for disaggregating census data
Gridded livestock distribution datasets have been produced for several years and are used in various fields, including epidemiology, livestock impact assessment, and territory management. Those datasets are based on census conducted at national/sub-national scale which are then downscaled using machine learning algorithm and relevant spatial-explicit environmental predictors. The most known dataset of livestock disaggregated observations is the Gridded Livestock of the World (GLW), which produces global maps of livestock density at 10 km spatial resolution for several livestock species. Though this spatial resolution can be appropriate to describe livestock distribution at the global scale, it inherently leads to a coarse representation of breeding species density for smaller territories such as the Caribbean Islands. In this study, we propose an adaptation of the GLW methodology that accounts for the spatial autocorrelation in observed cattle distribution, thereby better capturing the specific characteristics of geographically limited areas such as the Guadeloupean archipelago. Cattle census data were collected for the 32 municipalities of the archipelago and associated to environmental predictors derived from remote sensing and land cover datasets. Together with the Random Forest (RF) algorithm used in the standard GLW methodology, we tested the performance of a Geographical Random Forest (GRF), a novel methodology allowing for taking into account the spatial autocorrelation of the response variable. The GRF algorithm demonstrated significantly better performance compared to the RF algorithm, albeit with longer processing times, and allowed us producing cattle distribution maps for the entire Guadeloupe archipelago at a spatial resolution of 225 m using both algorithms. The approach developed holds potential for application to other small territories, including other islands in the Caribbean.
Open AccessDownscaling local distribution of cattle over Guadeloupe archipelago: an adapted method for disaggregating census data
Gridded livestock distribution datasets have been produced for several years and are used in various fields, including epidemiology, livestock impact assessment, and territory management. Those datasets are based on census conducted at national/sub-national scale which are then downscaled using machine learning algorithm and relevant spatial-explicit environmental predictors. The most known dataset of livestock disaggregated observations is the Gridded Livestock of the World (GLW), which produces global maps of livestock density at 10 km spatial resolution for several livestock species. Though this spatial resolution can be appropriate to describe livestock distribution at the global scale, it inherently leads to a coarse representation of breeding species density for smaller territories such as the Caribbean Islands. In this study, we propose an adaptation of the GLW methodology that accounts for the spatial autocorrelation in observed cattle distribution, thereby better capturing the specific characteristics of geographically limited areas such as the Guadeloupean archipelago. Cattle census data were collected for the 32 municipalities of the archipelago and associated to environmental predictors derived from remote sensing and land cover datasets. Together with the Random Forest (RF) algorithm used in the standard GLW methodology, we tested the performance of a Geographical Random Forest (GRF), a novel methodology allowing for taking into account the spatial autocorrelation of the response variable. The GRF algorithm demonstrated significantly better performance compared to the RF algorithm, albeit with longer processing times, and allowed us producing cattle distribution maps for the entire Guadeloupe archipelago at a spatial resolution of 225 m using both algorithms. The approach developed holds potential for application to other small territories, including other islands in the Caribbean.
- Open Access
- 1 December 2023
- Published by JLE in Environnement, Risques & Santé
- Vol. 22 (S1) , 64-72
Mapping cropping systems and associated practices to characterize pesticide pollution over time
To study the impact of pesticide uses in agriculture on human health, data on the types and quantities of pesticide active substances introduced into the environment over a long period of time are needed. However, when there is no systematic or available written records of pesticide applications, the characterization of environmental contamination remains challenging. In this technical note, we propose a method to organize heterogeneous information to represent the spatial dynamics of pesticide uses, with the case of Guadeloupe, French West Indies. Based on various data sources (paper maps, digital geographic data, articles and technical reports, pesticide products information, interviews), different databases were built: i) a land use database (1969-2019), with maps of the main three groups of crops in Guadeloupe: banana, sugar cane and market gardening; ii) a database of agricultural practices regarding the use of pesticides and their evolution over time for these groups of crops; iii) a database of available active substances according to time. Homogenized datasets of banana, sugar cane and market gardening plots allow to visualize the crop distribution at different periods of time and the evolution trend of cultivated areas. The database on agricultural practices regarding the use of pesticide includes the description of different 'pesticide itineraries'-characterized by information on the cropping system and on the schedule of pesticide applications, and their evolution through time. The database on active substances includes two types of information: for each year since 1960, the characteristics of authorized active substances based on regulation, and of available active substances, based on farmers' practices in Guadeloupe. All databases are made available on an open data repository; they will be used to derive indicators of pesticide pressure and study the link between cancer cases and the presence of pesticides in the environment.
- 1 December 2023
- Published by JLE in Environnement, Risques & Santé
- Vol. 22 (S1) , 64-72
Open AccessMapping cropping systems and associated practices to characterize pesticide pollution over time
To study the impact of pesticide uses in agriculture on human health, data on the types and quantities of pesticide active substances introduced into the environment over a long period of time are needed. However, when there is no systematic or available written records of pesticide applications, the characterization of environmental contamination remains challenging. In this technical note, we propose a method to organize heterogeneous information to represent the spatial dynamics of pesticide uses, with the case of Guadeloupe, French West Indies. Based on various data sources (paper maps, digital geographic data, articles and technical reports, pesticide products information, interviews), different databases were built: i) a land use database (1969-2019), with maps of the main three groups of crops in Guadeloupe: banana, sugar cane and market gardening; ii) a database of agricultural practices regarding the use of pesticides and their evolution over time for these groups of crops; iii) a database of available active substances according to time. Homogenized datasets of banana, sugar cane and market gardening plots allow to visualize the crop distribution at different periods of time and the evolution trend of cultivated areas. The database on agricultural practices regarding the use of pesticide includes the description of different 'pesticide itineraries'-characterized by information on the cropping system and on the schedule of pesticide applications, and their evolution through time. The database on active substances includes two types of information: for each year since 1960, the characteristics of authorized active substances based on regulation, and of available active substances, based on farmers' practices in Guadeloupe. All databases are made available on an open data repository; they will be used to derive indicators of pesticide pressure and study the link between cancer cases and the presence of pesticides in the environment.