Limits...
Ecological niche model of Phlebotomus alexandri and P. papatasi (Diptera: Psychodidae) in the Middle East.

Colacicco-Mayhugh MG, Masuoka PM, Grieco JP - Int J Health Geogr (2010)

Bottom Line: The bioclimatic and elevation variables all contributed to model development; however, none influenced the model as strongly as land cover.While not perfect representations of the absolute distribution of P. papatasi and P. alexandri, these models indicate areas with a higher probability of presence of these species.This information could be used to help guide future research efforts into the ecology of these species and epidemiology of the pathogens that they transmit.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Sand Fly Biology, Division of Entomology, Walter Reed Army Institute of Research, 503 Robert Grant Avenue, Silver Spring, MD 20910, USA. michelle.colacicco@us.army.mil

ABSTRACT

Background: The purpose of this study is to create distribution models of two sand fly species, Phlebotomus papatasi (Scopoli) and P. alexandri (Sinton), across the Middle East. Phlebotomus alexandri is a vector of visceral leishmaniasis, while P. papatasi is a vector of cutaneous leishmaniasis and sand fly fever. Collection records were obtained from literature reports from 1950 through 2007 and unpublished field collection records. Environmental layers considered in the model were elevation, precipitation, land cover, and WorldClim bioclimatic variables. Models were evaluated using the threshold-independent area under the curve (AUC) receiver operating characteristic analysis and the threshold-dependent minimum training presence.

Results: For both species, land cover was the most influential environmental layer in model development. The bioclimatic and elevation variables all contributed to model development; however, none influenced the model as strongly as land cover.

Conclusion: While not perfect representations of the absolute distribution of P. papatasi and P. alexandri, these models indicate areas with a higher probability of presence of these species. This information could be used to help guide future research efforts into the ecology of these species and epidemiology of the pathogens that they transmit.

Show MeSH

Related in: MedlinePlus

Jackknife test of training gain for P. alexandri. Environmental variables: bio1through bio 19 represent the bioclimatic variables (Table 2); xsub_alt is the elevation layer; xsub_g4 is the land cover layer.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2823717&req=5

Figure 4: Jackknife test of training gain for P. alexandri. Environmental variables: bio1through bio 19 represent the bioclimatic variables (Table 2); xsub_alt is the elevation layer; xsub_g4 is the land cover layer.

Mentions: As in the model for P. papatasi, land cover was the most influential variable in modeling P. alexandri (Figure 4). Jackknife tests show high training gain when land cover is considered alone and a large drop training gain when land cover is omitted from the model. The land cover types and probabilities associated with the training points are given in Table 1. As with P. papatasi, points classified as urban, field/woody savanna, and woody savanna coverages have high probabilities of presence. However, the sample sizes for each of these habitats are small. All other classes have either very wide ranges or small sample size.


Ecological niche model of Phlebotomus alexandri and P. papatasi (Diptera: Psychodidae) in the Middle East.

Colacicco-Mayhugh MG, Masuoka PM, Grieco JP - Int J Health Geogr (2010)

Jackknife test of training gain for P. alexandri. Environmental variables: bio1through bio 19 represent the bioclimatic variables (Table 2); xsub_alt is the elevation layer; xsub_g4 is the land cover layer.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2823717&req=5

Figure 4: Jackknife test of training gain for P. alexandri. Environmental variables: bio1through bio 19 represent the bioclimatic variables (Table 2); xsub_alt is the elevation layer; xsub_g4 is the land cover layer.
Mentions: As in the model for P. papatasi, land cover was the most influential variable in modeling P. alexandri (Figure 4). Jackknife tests show high training gain when land cover is considered alone and a large drop training gain when land cover is omitted from the model. The land cover types and probabilities associated with the training points are given in Table 1. As with P. papatasi, points classified as urban, field/woody savanna, and woody savanna coverages have high probabilities of presence. However, the sample sizes for each of these habitats are small. All other classes have either very wide ranges or small sample size.

Bottom Line: The bioclimatic and elevation variables all contributed to model development; however, none influenced the model as strongly as land cover.While not perfect representations of the absolute distribution of P. papatasi and P. alexandri, these models indicate areas with a higher probability of presence of these species.This information could be used to help guide future research efforts into the ecology of these species and epidemiology of the pathogens that they transmit.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Sand Fly Biology, Division of Entomology, Walter Reed Army Institute of Research, 503 Robert Grant Avenue, Silver Spring, MD 20910, USA. michelle.colacicco@us.army.mil

ABSTRACT

Background: The purpose of this study is to create distribution models of two sand fly species, Phlebotomus papatasi (Scopoli) and P. alexandri (Sinton), across the Middle East. Phlebotomus alexandri is a vector of visceral leishmaniasis, while P. papatasi is a vector of cutaneous leishmaniasis and sand fly fever. Collection records were obtained from literature reports from 1950 through 2007 and unpublished field collection records. Environmental layers considered in the model were elevation, precipitation, land cover, and WorldClim bioclimatic variables. Models were evaluated using the threshold-independent area under the curve (AUC) receiver operating characteristic analysis and the threshold-dependent minimum training presence.

Results: For both species, land cover was the most influential environmental layer in model development. The bioclimatic and elevation variables all contributed to model development; however, none influenced the model as strongly as land cover.

Conclusion: While not perfect representations of the absolute distribution of P. papatasi and P. alexandri, these models indicate areas with a higher probability of presence of these species. This information could be used to help guide future research efforts into the ecology of these species and epidemiology of the pathogens that they transmit.

Show MeSH
Related in: MedlinePlus