Content uploaded by Daniela Cianci
Author content
All content in this area was uploaded by Daniela Cianci on Oct 17, 2014
Content may be subject to copyright.
Faculty
of Veterinary
Medicine
Environmental data
•MODIS:
oMiddle Infra-red
oEnhanced vegetation index
oNormalised difference vegetation index
oDay-time land surface temperature
oNight-time land surface temperature
•Precipitation (CMORPH, Worldclim)
•MODIS Digital elevation model
•Population density
First attempt to model the spatial distribution of
mosquito species in the Netherlands using VecmapTM
Field data
• Dutch National Centre for Monitoring of
Vectors
• Cross-sectional study: weekly operated
CO2-baited traps, 766 catches
• April-October 2010-2013
1 Faculty of Veterinary Medicine, Utrecht University, Utrecht,
The Netherlands
Daniela Cianci1|Adolfo Ibáñez-Justicia2| Hans Heesterbeek1| Nienke Hartemink1
Departement
Name name name
Correspondence: d.cianci@uu.nl
This project was funded by EU
grant FP7-261504 EDENext
Modelling
• Non-linear
discriminant analysis
•Random forest
• Logistic regression
SPECIES ABSENCE PRESENCE
Culiseta annulata 320 446
Anopheles claviger 637 129 TFA: temporal
Fourier analysis
Sensitivity (Se), ability of a model to identify known positive sites. Specificity (Sp), ability of a model to identify known negative sites.
2 Dutch National Centre for Monitoring of Vectors, Food and
Consumer Product Safety Authority, Wageningen, The Netherlands
Discussion
• The maps show the predicted habitat suitability for Anopheles claviger and Culiseta annulata.
•For Culiseta annulata, the predictions are in agreement with the field experience. This species is
thought to be more present in the coastal areas of the Netherlands, to have a preference for inundated
and agricultural areas and to avoid forested areas.
•For Anopheles claviger, there were fewer presence points. The maps suggest that this species may
prefer areas with a clay soil, although this has not yet been reported in the literature.
ACKNOWLEDGEMENTS: David Morley, Els Ducheyne, Veerle Versteirt, Guy Hendrickx,
Neil Alexander, William Wint, Eva De Clercq, Sophie Vanwambeke
Non-linear
discriminant analysis
Field data Random forest Logistic regression
Se: 0.376
Sp: 0.892 Se: 0.613
Sp: 0.605 Se: 0.520
Sp: 0.694
Se: 0.453
Sp: 0.935 Se: 0.610
Sp: 0.560 Se: 0.658
Sp: 0.630
Vecmap™
http://iap.esa.int/projects/
health/vecmap