Search

Summary: Crowdsourcing methods facilitate the production of scientific information by non‐experts. This form of citizen science (CS) is becoming a key source of complementary data in many fields to inform data‐driven decisions and study challenging problems. However, concerns about the validity of these data often constrain their utility. In this paper, we focus on the use of citizen science data in addressing complex challenges in environmental conservation. We consider this issue from three perspectives. First, we present a literature scan of papers that have employed Bayesian models with citizen science in ecology. Second, we compare several popular majority vote algorithms and introduce a Bayesian item response model that estimates and accounts for participants' abilities after adjusting for the difficulty of the images they have classified. The model also enables participants to be clustered into groups based on ability. Third, we apply the model in a case study involving the classification of corals from underwater images from the Great Barrier Reef, Australia. We show that the model achieved superior results in general and, for difficult tasks, a weighted consensus method that uses only groups of experts and experienced participants produced better performance measures. Moreover, we found that participants learn as they have more classification opportunities, which substantially increases their abilities over time. Overall, the paper demonstrates the feasibility of CS for answering complex and challenging ecological questions when these data are appropriately analysed. This serves as motivation for future work to increase the efficacy and trustworthiness of this emerging source of data. [ABSTRACT FROM AUTHOR]

To model the price, price volatilities and their inter-relationships of the Australian wholesale spot electricity markets, the multivariate generalized autoregressive conditional heteroskedasticity (MGARCH) models have been applied (Higgs, Energy Economics 31 (5):748–56, 2009) and these models were extended to the sophisticated GARCH-vine-copula model (Apergis et al., Energy Economics 90:104834, 2020) recently. Stochastic volatility (SV) models, as flexible alternatives to GARCH models, have demonstrated their superiority in many financial applications. However, the use of SV models in the modeling of electricity markets is still quite limited. This paper investigates existing multivariate SV models and proposes efficient SV models with bilateral Granger causality and skew error distributions, to model the price and price volatilities of three pairs of markets, selected from four regional electricity markets in Australia, which are shown to be highly correlated in the Higgs (Energy Economics 31 (5):748–56, 2009) study. Bayesian approach using Markov chain Monte Carlo (MCMC) method is adopted and models are implemented using the software OpenBUGS. Based on Deviance Information Criterion, the models with skew error distributions perform better than those with symmetric distributions. [ABSTRACT FROM AUTHOR]

Introduction of pests and diseases through trade is one of the main socio‐ecological challenges worldwide. Although Binomial sampling inspection at the border can reduce pest entry risk, it is common for consignments to fail inspection, wasting resources for both exporter and importer. Outsourcing the inspection to the exporting country could reduce the cost of inspection for both parties. However, there is then a need to assess the quality of the offshore inspection. In this paper, we develop an inverse method combining past inspection data on the pathway, an onshore inspection sample, and the Beta‐Binomial model to infer the sample size of the offshore inspection. We illustrate the method on two case studies: the importation of live plants through germplasm into Australia and the importation of pelleted seeds in New Zealand. In these case studies, we found that detecting four to five infested units in a single onshore inspection was typically sufficient to significantly doubt the presence of a compliant offshore inspection. We also ran a simulation experiment to quantify the statistical power to reject or accept the presence of compliant offshore inspection in practice: In highly infested pathways, we could detect the absence of offshore inspections after inspecting five consignments onshore. Less infested pathways required inspecting 20 to 60 consignments onshore. Our study demonstrates that Binomial sampling onshore can be used to assess the quality of offshore inspections. [ABSTRACT FROM AUTHOR]

• Develops a deep Bayesian survival model that integrates deep neural network and Bayesian inference into the survival analysis for rail useful lifetime modelling. • Demonstrates survival analysis is an effective way to handle censored data problems in rail useful lifetime modelling. • Conducts extensive evaluation to verify the effectiveness of this model, using Concordance index, root mean square error, and a proposed CW-index. • The estimated lifetimes obtained by this model are the closest to real lifetimes compared to other commonly used models. • This model can provide the confidence interval of the estimated lifetime, which is more useful than point estimates in decision-making and maintenance planning of railroad systems. Reliable estimation of rail useful lifetime can provide valuable information for predictive maintenance in railway systems. However, in most cases, lifetime data is incomplete because not all pieces of rail experience failure by the end of the study horizon, a problem known as censoring. Ignoring or otherwise mistreating the censored cases might lead to false conclusions. Survival approach is particularly designed to handle censored data for analysing the expected duration of time until one event occurs, which is rail failure in this paper. This paper proposes a deep Bayesian survival approach named BNN-Surv to properly handle censored data for rail useful lifetime modelling. The proposed BNN-Surv model applies the deep neural network in the survival approach to capture the non-linear relationship between covariates and rail useful lifetime. To consider and quantify uncertainty in the model, Monte Carlo dropout, regarded as the approximate Bayesian inference, is incorporated into the deep neural network to provide the confidence interval of the estimated lifetime. The proposed approach is implemented on a four-year dataset including track geometry monitoring data, track characteristics data, various types of defect data, and maintenance and replacement (M&R) data collected from a section of railway tracks in Australia. Through extensive evaluation, including Concordance index (C-index) and root mean square error (RMSE) for evaluating model performance, as well as a proposed CW-index for evaluating uncertainty estimations, the effectiveness of the proposed approach is confirmed. The results show that, compared with other commonly used models, the proposed approach can achieve the best concordance index (C-index) of 0.80, and the estimated rail useful lifetimes are closer to real lifetimes. In addition, the proposed approach can provide the confidence interval of the estimated lifetime, with a correct coverage of 81% of the actual lifetime when the confidence interval is 1.38, which is more useful than point estimates in decision-making and maintenance planning of railroad systems. [ABSTRACT FROM AUTHOR]

In this study, Bayesian estimation is performed for a class of random coefficient self-exciting integer-valued threshold autoregressive processes with explanatory variables. A new model with a linear structure is obtained through model reconstruction, which makes Markov Chain Monte Carlo method easy to perform. By introducing the latent variables series, a complete data likelihood is obtained. Based on this likelihood, the full conditional distributions are easily obtained for all the parameters and latent variables. By maximizing the posterior probability function, the threshold parameter is accurately estimated. Finally, some numerical results of the estimates and a real data example of crime counts in Ballina, New South Wales, Australia are presented. [ABSTRACT FROM AUTHOR]

Monitoring waterbird populations in Australia is challenging for reasons of counting logistics, and because population aggregation and dispersion can shift rapidly in response to large spatio-temporal variations in resource availability. The East Australian Waterbird survey has conducted annual, aerial, systematic counts of waterbirds over eastern Australia for almost 40 years. It was designed to monitor waterbird populations using design-based inference though for many species this form of inference appears inadequate in the face of these challenges. Here we develop a state-space model-based Bayesian approach that, in addition to explicitly incorporating process noise and observation uncertainty, uses random effects and rainfall-derived covariates to model the year-to-year variation in the proportion of the total (super) population that is present on surveyed wetlands, and available to be counted. We use this model-based approach to estimate the superpopulation size of 45 waterbird species annually, and model the rate of population increase as a function of antecedent rainfall. The results confirm the strong positive effect of antecedent rainfall on population growth rates for nearly all species, and illustrate that species respond to rainfall differently in terms of habitat use, which influences whether they are present on surveyed wetlands. For many species, the year-to-year variation in the estimated proportion of the population on surveyed wetlands is very high. The results have implications for making inferences on population trends from these data, with the ability to model the year-to-year sampling variation a key requirement before the rate of population increase can be estimated with any precision. This study illustrates how to progress this approach, and infers that under average rainfall conditions, the general trend is for estimated superpopulation rates of increase to be negative, though for only a few species is this occurring with strong belief. [ABSTRACT FROM AUTHOR]

The redlegged earth mite, Halotydeus destructor (Tucker, 1925: Trombidiformes, Eupodoidea, Penthaleidae), is an invasive mite species. In Australia, this mite has become a pest of winter pastures and grain crops. We report the complete mitogenome for H. destructor, the first to represent the family Penthaleidae, superfamily Eupodoidea. The mitogenome of H. destructor is 14,691 bp in size, and has a GC content of 27.87%, 13 protein‐coding genes, two rRNA genes, and 22 tRNA genes. We explored evolutionary relationships of H. destructor with other members of the Trombidiformes using phylogenetic analyses of nucleotide sequences and the order of protein‐coding and rRNA genes. We found strong, consistent support for the superfamily Tydeoidea being the sister taxon to the superfamily Eupodoidea based on nucleotide sequences and gene arrangements. Moreover, the gene arrangements of Eupodoidea and Tydeoidea are not only identical to each other but also identical to that of the hypothesized arthropod ancestor, showing a high level of conservatism in the mitogenomic structure of these mite superfamilies. Our study illustrates the utility of gene arrangements for providing complementary information to nucleotide sequences with respect to inferring the evolutionary relationships of species within the order Trombidiformes. The mitogenome of H. destructor provides a valuable resource for further population genetic studies of this important agricultural pest. Given the co‐occurrence of closely related, morphologically similar Penthaleidae mites with H. destructor in the field, a complete mitogenome provides new opportunities to develop metabarcoding tools to study mite diversity in agro‐ecosystems. Moreover, the H. destructor mitogenome fills an important taxonomic gap that will facilitate further study of trombidiform mite evolution. [ABSTRACT FROM AUTHOR]