RESUMO
This study contributes a first comparison of current and potential threats to Natural World Heritage Sites from climate change, as assessed by experts, when site and location characteristics (size, year of inscription to the World Heritage list, continent, climate zone and kind of site) are controlled for. The probability of a threat as well as its intensity is analysed. Another novelty lies in the use of data from the IUCN Conservation Outlook Assessment, covering all 245 Natural and Mixed World Heritage Sites across the world for three points in time: 2014, 2017 and 2020. The threat of climate change is broadly defined and includes temperature extremes, rising temperatures, disappearing glaciers, coral bleaching, droughts, desertification, and rising sea levels. Results based on a simultaneous Probit model with random effects show that the probability of actual and potential climate change threats increases over time, but with differences for size, kind of site and location. The probability that a threat is identified is highest for marine and coastal sites, and for those in Latin America, while it is significantly lower for sites on the African continent. Larger sites have a higher probability of being assessed as at risk and the severity of threats is found to be lower for recently inscribed sites. The rate at which the likelihood of a threat assessment increases is consistent for both current and future situations, while the probability of the most severe threat is larger for the current than the future period. A serious threat from climate change is assessed as highest for locations in the tropical monsoon (current period) or the tropical savannah climate (future period). Estimations also show that pure descriptive statistics or bivariate correlations may not correctly identify the risk or the dignity of a threat.
RESUMO
BACKGROUND: Dengue is a wide-spread mosquito-borne disease globally with a likelihood of becoming endemic in tropical Queensland, Australia. The aim of this study was to analyse the spatial variation of dengue notifications in relation to climate variability and socio-ecological factors in the tropical climate zone of Queensland, Australia. METHODS: Data on the number of dengue cases and climate variables including minimum temperature, maximum temperature and rainfall for the period of January 1st, 2010 to December 31st, 2015 were obtained for each Statistical Local Area (SLA) from Queensland Health and Australian Bureau of Meteorology, respectively. Socio-ecological data including estimated resident population, percentage of Indigenous population, housing structure (specifically terrace house), socio-economic index and land use types for each SLA were obtained from Australian Bureau of Statistics, and Australian Bureau of Agricultural and Resource Economics and Sciences, respectively. To quantify the relationship between dengue, climate and socio-ecological factors, multivariate Poisson regression models in a Bayesian framework were developed with a conditional autoregressive prior structure. Posterior parameters were estimated using Bayesian Markov Chain Monte Carlo simulation with Gibbs sampling. RESULTS: In the tropical climate zone of Queensland, the estimated number of dengue cases was predicted to increase by 85% [95% Credible Interval (CrI): 25%, 186%] and 7% (95% CrI: 0.1%, 14%) for a 1-mm increase in average annual rainfall and 1% increase in the proportion of terrace houses, respectively. The estimated spatial variation (structured random effects) was small compared to the remaining unstructured variation, suggesting that the inclusion of covariates resulted in no residual spatial autocorrelation in dengue data. CONCLUSIONS: Climate and socio-ecological factors explained much of the heterogeneity of dengue transmission dynamics in the tropical climate zone of Queensland. Results will help to further understand the risk factors of dengue transmission and will provide scientific evidence in designing effective local dengue control programs in the most needed areas.