Abstract
The survival of ectothermic species is heavily dependent on environmental conditions, such as temperature and water balance. Understanding their survival responses to abiotic factors could help predict impacts of climate change on their population dynamics and human health. However, making a statistical inference and formulating a predictive model for mortality rates can be challenging when the observation numbers are limited. This study proposed an expert opinion elicitation framework that integrates expert opinions as prior distributions for the effects of continuous explanatory variables, through a Bayesian Parametric Survival Model (B-PSM). A historical survival dataset of female Ixodes ricinus ticks (Acari: Ixodidae) with small sample size was used. A total of 6 acarologists were recruited as experts for interactive online interview sessions to provide their opinions on average survival time under 4 different temperature and humidity scenarios. Most experts shared similar opinions on the effects of abiotic variables, and none of the experts was confident in the interaction effect. The variation of the opinions across multiple experts was handled by two approaches: 1) pooling and 2) averaging methods. The results showed that the pooling approach retains the variations of expert opinions, it may also disregard some irrelevant opinions to the observed data. While the averaging approach forms a numerical consensus across all the experts, but it may be less informative when the opinions distinctly diverge. The survival time of I. ricinus was found to be best described by the Weibull distribution, suggesting the mortality rate of ticks increases over time (aging effects). Also, the posterior predictions revealed that I. ricinus ticks were susceptible to desiccation conditions, with an interaction effect with the temperature. Therefore, our results suggested that relative humidity is an important factor in the survival of I. ricinus that should not be disregarded when evaluating the impacts of climate change on their population dynamics. Finally, this study provided a guideline for implementing the B-PSM framework to incorporate expert opinions and develop predictive survival models that can be applied in other ecological contexts.
Highlights
Survival is a fundamental ecological process that defines the de mographics of a population
The Akaike information criterion (AIC) values of survival regression models assuming a Weibull distribution suggested that the survival time T could be explained by a combination of non-linear effect of relative humidity, liner effect of temperature, and their interaction (Model B11; AIC 815.54; Table S3)
The present study proposed a hierarchical Bayesian parametric survival modeling (B-Parametric Survival Model (PSM)) framework that incorporates expert opinions as supplementary information on temperature and relative humidity effects on the survival time of female I. ricinus ticks
Summary
Survival is a fundamental ecological process that defines the de mographics of a population. In the face of anthropogenic climate change, the survival of invertebrate ectothermic species, such as insects and acari, has been considerably impacted by long-term alteration in abiotic conditions. Climate change may alter the distribution and community composition of ectothermic species globally, potentially affecting human well-being (Pecl et al, 2017). Climate change-induced alteration in the population dynamics of pollinators, such as bumblebees, and arthropod vectors, such as mosquitoes and ticks, may have had a negative impact on food security (Giannini et al, 2017) and infectious disease distributions (Dumic and Severnini, 2018; Lee et al, 2018), respectively. Understanding survival responses of invertebrate ecto therms to abiotic factors, the interaction between thermo-regulation and hydro-regulation (Rozen-Rechels et al, 2019), could help predict climate change impacts on their population dynamics and human health (Nadeau et al, 2017)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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 call
