Abstract
BackgroundDengue dynamics result from the complex interactions between the virus, the host and the vector, all being under the influence of the environment. Several studies explored the link between weather and dengue dynamics and some investigated the impact of climate change on these dynamics. Most attempted to predict incidence rate at a country scale or assess the environmental suitability at a global or regional scale. Here, we propose a new approach which consists in modeling the risk of dengue outbreak at a local scale according to climate conditions and study the evolution of this risk taking climate change into account. We apply this approach in New Caledonia, where high quality data are available.MethodsWe used a statistical estimation of the effective reproduction number (Rt) based on case counts to create a categorical target variable : epidemic week/non-epidemic week. A machine learning classifier has been trained using relevant climate indicators in order to estimate the probability for a week to be epidemic under current climate data and this probability was then estimated under climate change scenarios.ResultsWeekly probability of dengue outbreak was best predicted with the number of days when maximal temperature exceeded 30.8°C and the mean of daily precipitation over 80 and 60 days prior to the predicted week respectively. According to scenario RCP8.5, climate will allow dengue outbreak every year in New Caledonia if the epidemiological and entomological contexts remain the same.ConclusionWe identified locally relevant climatic factor driving dengue outbreaks in New Caledonia and assessed the inter-annual and seasonal risk of dengue outbreak under different climate change scenarios up to the year 2100. We introduced a new modeling approach to estimate the risk of dengue outbreak depending on climate conditions. This approach is easily reproducible in other countries provided that reliable epidemiological and climate data are available.
Highlights
Dengue dynamics result from the complex interactions between the virus, the host and the vector, all being under the influence of the environment
In contrary with machine learning algorithms, given their empirical nature, biological processes involved in the relation between features and the target variable are implicit. In this present study we introduced a complete method to forecast the current weekly probability of dengue outbreak for a specific location (e.g. New Caledonia) based on climate variability and estimate the inter-annual and seasonal risk of dengue outbreak during the century facing climate change
After investigating the inter-annual risk of dengue outbreak, we looked at the evolution of the profile of the seasonal risk of dengue outbreak
Summary
Dengue dynamics result from the complex interactions between the virus, the host and the vector, all being under the influence of the environment. Most studies have focused on predicting dengue incidence in the future and seldom studies have projected probability of dengue outbreak or risk of dengue outbreak for a specific location in the face of climate change [9,10,11,12]. The latter relied on mechanistic models that use simplified mathematical formulation of biological processes to make predictions. In contrary with machine learning algorithms, given their empirical nature, biological processes involved in the relation between features and the target variable are implicit
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
