PT  - JOURNAL ARTICLE
AU  - Micanaldo Ernesto Francisco
AU  - Thaddeus M. Carvajal
AU  - Masahiro Ryo
AU  - Kei Nukazawa
AU  - Divina M. Amalin
AU  - Kozo Watanabe
TI  - Dengue Disease Dynamics are Modulated by the Combined Influence of Precipitation and Landscapes: A Machine Learning-based Approach
AID  - 10.1101/2020.09.01.278713
DP  - 2020 Jan 01
TA  - bioRxiv
PG  - 2020.09.01.278713
4099  - http://biorxiv.org/content/early/2020/09/03/2020.09.01.278713.short
4100  - http://biorxiv.org/content/early/2020/09/03/2020.09.01.278713.full
AB  - Background Dengue is an endemic vector-borne disease influenced by environmental factors such as landscape and climate. Previous studies separately assessed the effects of landscape and climate factors on mosquito occurrence and dengue incidence. However, both factors interact in time and space to affect mosquito development and dengue disease transmission. For example, eggs laid in a suitable environment can hatch after being submerged in rain or flood water.Objectives This study aimed to investigate the combined influences of landscape and climate factors on mosquito occurrence and dengue incidence.Methods Entomological, epidemiological, and landscape data from the rainy season (July-December) were obtained from respective government agencies in Metro Manila, Philippines, from 2012 to 2014. Temperature, precipitation, and vegetation data were obtained through remote sensing. A random forest algorithm was used to select the landscape and climate variables. Afterwards, using the identified key variables, a model-based (MOB) recursive partitioning was implemented to test the combinatory influences of landscape and climate factors on the ovitrap index and dengue incidence.Results The MOB recursive partitioning for the ovitrap index indicated that mosquito occurrence was higher in high residential density areas, where industrial areas also exist and are well connected with roads. Precipitation was another key covariate modulating the effects of landscape factors, possibly by expanding breeding sites and activating mosquito reproduction. Moreover, the MOB recursive partitioning indicated that precipitation was the main predictor of dengue incidence, with a stronger effect in high residential density and commercial areas.Discussion Precipitation with floods has epidemiologically important implications by damaging shelters and causing population displacement, thus increasing exposure to dengue vectors. Our findings suggest that the intensification of vector control during the rainy season can be prioritized in residential and commercial areas to better control dengue disease dynamics.Competing Interest StatementThe authors have declared no competing interest.