Modelling adult Aedes aegypti and Aedes albopictus survival at different temperatures in laboratory and field settings

Oliver J. Brady, Michael A. Johansson, Carlos A. Guerra, Samir Bhatt, Nick Golding, David M. Pigott, Hélène Delatte, Marta G. Grech, Paul T. Leisnham, Rafael Maciel‐de‐Freitas, Linda M. Styer, David L. Smith, Thomas W. Scott, Peter W. Gething, Simon I HayParasites 2013 · DOI

Abstract

Our results support the importance of producing site-specific mosquito survival estimates. By including fluctuating temperature regimes, our models provide insight into seasonal patterns of Ae. aegypti and Ae. albopictus population dynamics that may be relevant to seasonal changes in dengue virus transmission. Our models can be integrated with Aedes and dengue modelling efforts to guide and evaluate vector control, better map the distribution of disease and produce early warning systems for dengue epidemics.

Introduction: The Lab-Field Gap

Predicting the spread of mosquito-borne diseases like dengue depends heavily on one critical variable: the lifespan of an adult female mosquito. A longer life translates directly into more opportunities to bite humans, more time for a virus to incubate, and increased chances of transmission.

However, a significant "lab-field gap" persists in entomological research. Laboratory studies offer precise data under controlled, constant temperatures without predators, but they fail to reflect the harsh realities of the wild. Conversely, field studies using Mark-Release-Recapture (MRR) techniques are realistic but often sparse, noisy, and limited to narrow temperature ranges. This research reconciles these two data sources by applying Generalized Additive Models (GAMs) to a massive dataset of 410 experiments, producing the first comprehensive models that predict survival across the full range of temperatures encountered in the wild.

Methodology

To overcome the limitations of traditional parametric models (such as Gompertz or Weibull functions), which often force biological data into rigid, unrealistic shapes, the researchers utilized Generalized Additive Models (GAMs). The workflow involved several key stages to bridge the gap between laboratory precision and field reality.

Figure 3: Workflow for developing the adjusted temperature survival model, integrating laboratory longevity data with field MRR data.
  1. Lab Modeling: GAMs were first fitted to laboratory data to capture the complex, non-linear relationships between temperature, age, and survival.
  2. Field Reconciliation: These lab models were "adjusted" using field data. By comparing lab-predicted survival to observed field survival, the researchers calculated "external mortality"—death caused by predation, weather, and other environmental hazards.
  3. Fluctuation Simulation: Rather than relying on constant mean temperatures, the model simulated daily fluctuating temperature regimes (sinusoidal cycles) to mirror real-world conditions.
  4. Uncertainty Propagation: Bootstrapping with 200 repeats ensured that the uncertainty inherent in sparse field data was accurately reflected in the final confidence intervals.

Model Selection and Performance

Traditional parametric models often struggle to maintain accuracy across the entire thermal range. Analysis of the relative likelihood of various parametric models demonstrated that no single function (Log-logistic, Gompertz, Exponential, or Weibull) was universally superior across all temperatures for either species.

Figure 5: Relative likelihood of four parametric models for Ae. aegypti and Ae. albopictus across a range of temperatures, illustrating the inconsistency of traditional fits.

By moving to GAMs, the researchers were able to capture 91.1% of the variation in Ae. aegypti laboratory data. This approach successfully incorporated "critical thermal limits," such as the sharp drop in survival for Ae. aegypti below 14-15°C, providing a robust biological basis for the geographical limits of these vectors.

Key Insights

  • The Dominance of External Mortality: In the field, "accidental" death is so prevalent that it dwarfs the effects of biological aging (senescence). For Ae. aegypti, field mortality was found to be 4.5 times higher than laboratory mortality. While mosquitoes do age biologically, the high external mortality rate means few individuals in the wild live long enough for senescence to become the primary cause of death.
  • Species-Specific Thermal Profiles: Ae. albopictus generally lives longer than Ae. aegypti in both settings and maintains high survival across a broad plateau (20°C to 30°C). In contrast, Ae. aegypti survival peaks sharply around 21°C but is more robust to extreme temperatures, tolerating a wider viable range overall.

Conclusion and Future Directions

While these models represent a significant advancement, limitations remain, particularly the lack of field survival data at extreme temperatures. Most MRR studies are conducted in moderate tropical or subtropical conditions, leaving gaps in our understanding of survival at the thermal edges. Future work may utilize transcript-level aging markers to better validate the "tail end" of the survival curve in the wild. Ultimately, these models are intended for integration into early-warning systems, allowing public health officials to predict dengue transmission risk based on local weather forecasts and seasonal population dynamics.

Summary updated

Abstract

Our results support the importance of producing site-specific mosquito survival estimates. By including fluctuating temperature regimes, our models provide insight into seasonal patterns of Ae. aegypti and Ae. albopictus population dynamics that may be relevant to seasonal changes in dengue virus transmission. Our models can be integrated with Aedes and dengue modelling efforts to guide and evaluate vector control, better map the distribution of disease and produce early warning systems for dengue epidemics.
BibTeX
@inproceedings{brady2013modelling,
  author = {Oliver J. Brady and Michael A. Johansson and Carlos A. Guerra and Samir Bhatt and Nick Golding and David M. Pigott and Hélène Delatte and Marta G. Grech and Paul T. Leisnham and Rafael Maciel‐de‐Freitas and Linda M. Styer and David L. Smith and Thomas W. Scott and Peter W. Gething and Simon I Hay},
  title = {{Modelling adult Aedes aegypti and Aedes albopictus survival at different temperatures in laboratory and field settings}},
  booktitle = {Parasites},
  year = {2013},
  doi = {10.1186/1756-3305-6-351},
}
Figure