Background: Mosquito-borne diseases are a growing concern for temperate regions including the northeastern US. There the two primary mosquito vectors, Cx. pipiens and Ae. albopictus are widespread, endemic circulation of West Nile virus causes sporadic outbreaks, and imported arboviruses such as dengue, chikungunya, and Zika are on the rise. With temperate mosquito-borne disease outbreaks likely to increase in frequency, it is critical to reduce mosquito populations in the northeastern US. Community-based source reduction is heralded as the most sustainable component of integrated mosquito management. Yet mosquitoes develop rapidly, requiring weekly maintenance of mosquito habitat. This is onerous and community commitment flags. The development of predictive models to inform focused vector-control efforts is therefore of great utility.
Objectives and Methods: The overarching objective of this research is to make robust predictive modeling frameworks based on empirically derived relationships of the ecology and epidemiology of mosquito-borne disease systems in the northeastern US. We aim to quantify the relationships between local environmental and meteorological conditions and mosquito vectors. In Chapters 2 and 4 we use lengthy surveillance records to develop models and use model ensembles to generate predictions based on out-of-sample data. For chapter 3 we use more spatially refined data to investigate the influence of intra-urban heterogeneities and how climatic conditions influence mosquito populations across these defined differences.
Results: In Chapter 2, we model and forecast WNV infection rates among mosquito vectors using meteorological and hydrological conditions. We show that real-time climate information can predict WNV Culex infection rates prior to when human risk is greatest. In Chapter 3, we link infrastructure degradation and vegetation patterns with Ae. albopictus infestation levels as well as the interactive effect of precipitation across these environmental conditions. In Chapter 4, we identify key land use characteristics and meteorological conditions associated with annual Ae. albopictus abundance. Further we use imported chikungunya cases to delineate areas of high arboviral importation and, in combination with areas of high Ae. albopictus abundance, areas at heightened risk for arboviral transmission.
Conclusions: While temperate outbreaks are often self-limiting they may be increasing in frequency and severity. Due to the multitude of invasive vectors and arboviruses, vector control techniques that work for multiple mosquito species are likely more effective and sustainable. Here we build build empirical models that accurately predict mosquito dynamics before populations peak which is critical for vector control. We recommend integrating predictive modeling into mosquito management guidelines as this could focus valuable resources to when and where mosquito-borne transmission risk is greatest. Further we find social and ecological determinants of mosquito dynamics, supporting further study that combine socio-ecological processes into model frameworks.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/D8MC9B90 |
| Date | January 2017 |
| Creators | Little, Eliza Anastazia Hazel |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.002 seconds