1
|
Bon JJ, Bretherton A, Buchhorn K, Cramb S, Drovandi C, Hassan C, Jenner AL, Mayfield HJ, McGree JM, Mengersen K, Price A, Salomone R, Santos-Fernandez E, Vercelloni J, Wang X. Being Bayesian in the 2020s: opportunities and challenges in the practice of modern applied Bayesian statistics. PHILOSOPHICAL TRANSACTIONS. SERIES A, MATHEMATICAL, PHYSICAL, AND ENGINEERING SCIENCES 2023; 381:20220156. [PMID: 36970822 PMCID: PMC10041356 DOI: 10.1098/rsta.2022.0156] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 08/22/2022] [Accepted: 01/06/2023] [Indexed: 06/18/2023]
Abstract
Building on a strong foundation of philosophy, theory, methods and computation over the past three decades, Bayesian approaches are now an integral part of the toolkit for most statisticians and data scientists. Whether they are dedicated Bayesians or opportunistic users, applied professionals can now reap many of the benefits afforded by the Bayesian paradigm. In this paper, we touch on six modern opportunities and challenges in applied Bayesian statistics: intelligent data collection, new data sources, federated analysis, inference for implicit models, model transfer and purposeful software products. This article is part of the theme issue 'Bayesian inference: challenges, perspectives, and prospects'.
Collapse
Affiliation(s)
- Joshua J. Bon
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Adam Bretherton
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Katie Buchhorn
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Susanna Cramb
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Public Health and Social Work, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Christopher Drovandi
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Conor Hassan
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Adrianne L. Jenner
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Helen J. Mayfield
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Public Health, The University of Queensland, Saint Lucia, Queensland, Australia
| | - James M. McGree
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Kerrie Mengersen
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Aiden Price
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Robert Salomone
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Computer Science, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Edgar Santos-Fernandez
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Julie Vercelloni
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
| | - Xiaoyu Wang
- Centre for Data Science, Queensland University of Technology, Brisbane, Queensland, Australia
- School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
| |
Collapse
|
2
|
Nolan V, Gilbert F, Reed T, Reader T. Distribution models calibrated with independent field data predict two million ancient and veteran trees in England. ECOLOGICAL APPLICATIONS : A PUBLICATION OF THE ECOLOGICAL SOCIETY OF AMERICA 2022; 32:e2695. [PMID: 35732507 PMCID: PMC10078183 DOI: 10.1002/eap.2695] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 02/26/2021] [Revised: 02/25/2022] [Accepted: 03/31/2022] [Indexed: 06/15/2023]
Abstract
Large, citizen-science species databases are powerful resources for predictive species distribution modeling (SDM), yet they are often subject to sampling bias. Many methods have been proposed to correct for this, but there exists little consensus as to which is most effective, not least because the true value of model predictions is hard to evaluate without extensive independent field sampling. We present here a nationwide, independent field validation of distribution models of ancient and veteran trees, a group of organisms of high conservation importance, built using a large and internationally unique citizen-science database: the Ancient Tree Inventory (ATI). This validation exercise presents an opportunity to test the performance of different methods of correcting for sampling bias, in the search for the best possible prediction of ancient and veteran tree distributions in England. We fitted a variety of distribution models of ancient and veteran tree records in England in relation to environmental predictors and applied different bias correction methods, including spatial filtering, background manipulation, the use of bias files, and, finally, zero-inflated (ZI) regression models, a new method with great potential to investigate and remove sampling bias in species data. We then collected new independent field data through systematic surveys of 52 randomly selected 1-km2 grid squares across England to obtain abundance estimates of ancient and veteran trees. Calibration of the distribution models against the field data suggests that there are around eight to 10 times as many ancient and veteran trees present in England than the records currently suggest, with estimates ranging from 1.7 to 2.1 million trees compared to the 200,000 currently recorded in the ATI. The most successful bias correction method was systematic sampling of occurrence records, although the ZI models also performed well, significantly predicting field observations and highlighting both likely causes of undersampling and areas of the country in which many unrecorded trees are likely to be found. Our findings provide the first robust nationwide estimate of ancient and veteran tree abundance and demonstrate the enormous potential for distribution modeling based on citizen-science data combined with independent field validation to inform conservation planning.
Collapse
Affiliation(s)
| | | | | | - Tom Reader
- Life SciencesUniversity of NottinghamNottinghamUK
| |
Collapse
|
3
|
Sun CC, Hurst JE, Fuller AK. Citizen Science Data Collection for Integrated Wildlife Population Analyses. Front Ecol Evol 2021. [DOI: 10.3389/fevo.2021.682124] [Citation(s) in RCA: 6] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/02/2023] Open
Abstract
Citizen science, or community science, has emerged as a cost-efficient method to collect data for wildlife monitoring. To inform research and conservation, citizen science sampling designs should collect data that match the robust statistical analyses needed to quantify species and population patterns. Further increasing the contributions of citizen science, integrating citizen science data with other datasets and datatypes can improve population estimates and expand the spatiotemporal extent of inference. We demonstrate these points with a citizen science program called iSeeMammals developed in New York state in 2017 to supplement costly systematic spatial capture-recapture sampling by collecting opportunistic data from one-off observations, hikes, and camera traps. iSeeMammals has initially focused on the growing population of American black bear (Ursus americanus), with integrated analysis of iSeeMammals camera trap data with systematic data for a region with a growing bear population. The triumvirate of increased spatial and temporal coverage by at least twofold compared to systematic sampling, an 83% reduction in annual sampling costs, and improved density estimates when integrated with systematic data highlight the benefits of collecting presence-absence data in citizen science programs for estimating population patterns. Additional opportunities will come from applying presence-only data, which are oftentimes more prevalent than presence-absence data, to integrated models. Patterns in data submission and filtering also emphasize the importance of iteratively evaluating patterns in engagement, usability, and accessibility, especially focusing on younger adult and teenage demographics, to improve data quality and quantity. We explore how the development and use of integrated models may be paired with citizen science project design in order to facilitate repeated use of datasets in standalone and integrated analyses for supporting wildlife monitoring and informing conservation.
Collapse
|
4
|
Matthews RW. Patterns and composition of medium and large vertebrate roadkill, based on six annual surveys along two adjoining highways in south-eastern Queensland, Australia. AUSTRALIAN MAMMALOGY 2020. [DOI: 10.1071/am19044] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/23/2022]
Abstract
Six annual single-pass roadkill surveys along two adjoining rural Queensland highways near Carnarvon Gorge National Park revealed 612 medium-size to large vertebrates, representing more than 18 taxa. Most were mammals (92%), particularly macropods. Losses averaged 0.26 animals km–1 year–1 (range = 0.17–0.33), with variation possibly reflecting road repair/reconstruction and record seasonal rainfalls. Annual roadkill totals for the 390-km highway were projected to be over 5000 vertebrates, with more than half being large macropods. A consistent hotspot or ecological trap was noted along a 17-km high-traffic-volume stretch north of Roma. Because the sparsely populated outback is habitat for much Australian wildlife, multiyear baseline data are vital to identify the magnitude of the problem and inform future research.
Collapse
|
5
|
Shima AL, Gillieson DS, Crowley GM, Dwyer RG, Berger L. Factors affecting the mortality of Lumholtz's tree-kangaroo (Dendrolagus lumholtzi) by vehicle strike. WILDLIFE RESEARCH 2018. [DOI: 10.1071/wr17143] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/23/2022]
Abstract
Context
Vehicle strike is a major issue where wildlife habitat is intersected by busy roads. Near Threatened Lumholtz’s tree-kangaroo (Dendrolagus lumholtzi) is a large (5–10 kg) semi-arboreal mammal found in populated rural and forested areas of north-eastern Australia. Warning signs, rope bridges and underpasses have not prevented ~20 animals being killed on the road each year.
Aims
To identify factors influencing Lumholtz’s tree-kangaroo vehicle strike to help inform mitigation options.
Methods
Citizen sightings (1998–2000) and 90 road-kills collected over 4.5 years on the Atherton Tablelands, Australia, were examined to determine the causes of vehicle strike in Lumholtz’s tree-kangaroo. The spatial distributions of sightings and road-kills were characterised using nearest-neighbour analysis, and the relationship between them was determined using a Bayesian approach that accounted for spatial autocorrelation. Gender, age, weight, season, rainfall, road and verge characteristics, traffic volumes, speed limits and mitigation measures were recorded to assess their influence on road-kill risk. Adequacy of speed limits to prevent collisions along road sections with more than four road-kills per 8 km (hazard zones) was assessed from visibility and stopping distances.
Key results
Vehicle strikes mainly affected male tree-kangaroos (2–5 years, 5.5–8 kg), occurred where live animals were most frequently sighted and were most likely on roads with narrow verges, low visibility and medium traffic volumes. Speed limits at hazard zones were inadequate to prevent collisions. Few warning signs corresponded with these zones, and road mortalities persisted where they did.
Conclusions
Unpredictable dispersal of young males and vehicle speeds unsuited to road conditions drive road mortalities in Lumholtz’s tree-kangaroo. Because tree-kangaroos do not appear to respond to existing mitigation measures, reducing traffic speeds, and increasing visibility, appear to be the most effective mitigation strategies for reducing tree-kangaroo road mortality.
Implications
Our findings suggest that tree-kangaroo road-kill can be reduced by reducing speed limits in line with government recommendations and increasing visibility by clearing road verges along sections of road with the highest tree-kangaroo mortality. Warning signage should be re-evaluated to determine whether its effectiveness can be improved.
Collapse
|
6
|
Santori C, Spencer RJ, Van Dyke JU, Thompson MB. Road mortality of the eastern long-necked turtle (Chelodina longicollis) along the Murray River, Australia: an assessment using citizen science. AUST J ZOOL 2018. [DOI: 10.1071/zo17065] [Citation(s) in RCA: 15] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/23/2022]
Abstract
Turtles face a variety of threats (e.g. habitat destruction, introduced predators) that are pushing many species towards extinction. Vehicle collisions are one of the main causes of mortality of adult freshwater turtles. To conceptualise the level of threat that roads pose to Australians turtles, we analysed data gathered through the citizen science project TurtleSAT along the Murray River. We recorded 124 occurrences of turtle road mortality, which included all three local species (Chelodina expansa, Chelodina longicollis, and Emydura macquarii). Chelodina longicollis was the most commonly reported species killed on roads. We found that rain and time of year affect the likelihood of C. longicollis being killed on roads: increased turtle mortality is associated with rain events and is highest during the month of November, which coincides with their nesting season. Chelodina longicollis was most likely to be killed on the Hume Highway and roads around major urban centres; therefore, we recommend that governing bodies focus management practices and increase awareness at these locations. The degree of road mortality that we detected in this study requires mitigation, as it may contribute to the decline of C. longicollis along the Murray River.
Collapse
|
7
|
Visintin C, van der Ree R, McCarthy MA. Consistent patterns of vehicle collision risk for six mammal species. JOURNAL OF ENVIRONMENTAL MANAGEMENT 2017; 201:397-406. [PMID: 28704730 DOI: 10.1016/j.jenvman.2017.05.071] [Citation(s) in RCA: 14] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/31/2016] [Revised: 05/05/2017] [Accepted: 05/23/2017] [Indexed: 06/07/2023]
Abstract
The occurrence and rate of wildlife-vehicle collisions are related to both anthropocentric and environmental variables, however, few studies compare collision risks for multiple species within a model framework that is adaptable and transferable. Our research compares collision risk for multiple species across a large geographic area using a conceptually simple risk framework. We used six species of native terrestrial mammal often involved with wildlife-vehicle collisions in south-east Australia. We related collisions reported to a wildlife organisation to the co-occurrence of each species and a threatening process (presence and movement of road vehicles). For each species, we constructed statistical models from wildlife atlas data to predict occurrence across geographic space. Traffic volume and speed on road segments (also modelled) characterised the magnitude of threatening processes. The species occurrence models made plausible spatial predictions. Each model reduced the unexplained variation in patterns and distributions of species between 29.5% (black wallaby) and 34.3% (koala). The collision models reduced the unexplained variation in collision event data between 7.4% (koala) and 19.4% (common ringtail possum) with predictor variables correlating similarly with collision risk across species. Road authorities and environmental managers need simple and flexible tools to inform projects. Our model framework is useful for directing mitigation efforts (e.g. on road effects or species presence), predicting risk across differing spatial and temporal scales and target species, inferring patterns of threat, and identifying areas warranting additional data collection, analysis, and study.
Collapse
Affiliation(s)
- Casey Visintin
- Quantitative and Applied Ecology Group, School of BioSciences, University of Melbourne, Parkville, VIC, 3010, Australia.
| | - Rodney van der Ree
- School of BioSciences, University of Melbourne, Parkville, VIC, 3010, Australia; Ecology and Infrastructure International Pty Ltd, PO Box 6031, Wantirna, VIC, 3152, Australia.
| | - Michael A McCarthy
- Quantitative and Applied Ecology Group, School of BioSciences, University of Melbourne, Parkville, VIC, 3010, Australia.
| |
Collapse
|
8
|
Madliger CL, Franklin CE, Hultine KR, van Kleunen M, Lennox RJ, Love OP, Rummer JL, Cooke SJ. Conservation physiology and the quest for a 'good' Anthropocene. CONSERVATION PHYSIOLOGY 2017; 5:cox003. [PMID: 28852507 PMCID: PMC5570019 DOI: 10.1093/conphys/cox003] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 11/27/2016] [Revised: 12/31/2016] [Accepted: 01/06/2017] [Indexed: 05/21/2023]
Abstract
It has been proposed that we are now living in a new geological epoch known as the Anthropocene, which is specifically defined by the impacts that humans are having on the Earth's biological diversity and geology. Although the proposal of this term was borne out of an acknowledgement of the negative changes we are imparting on the globe (e.g. climate change, pollution, coastal erosion, species extinctions), there has recently been action amongst a variety of disciplines aimed at achieving a 'good Anthropocene' that strives to balance societal needs and the preservation of the natural world. Here, we outline ways that the discipline of conservation physiology can help to delineate a hopeful, progressive and productive path for conservation in the Anthropocene and, specifically, achieve that vision. We focus on four primary ways that conservation physiology can contribute, as follows: (i) building a proactive approach to conservation; (ii) encouraging a pragmatic perspective; (iii) establishing an appreciation for environmental resilience; and (iv) informing and engaging the public and political arenas. As a collection of passionate individuals combining theory, technological advances, public engagement and a dedication to achieving conservation success, conservation physiologists are poised to make meaningful contributions to the productive, motivational and positive way forward that is necessary to curb and reverse negative human impact on the environment.
Collapse
Affiliation(s)
- Christine L. Madliger
- Fish Ecology and Conservation Physiology Laboratory, Department of Biology and Institute of Environmental Science, Carleton University, 1125 Colonel By Drive, Ottawa, ON, CanadaK1S 5B6
- Department of Biological Sciences, University of Windsor, 401 Sunset Avenue, ON, CanadaN9B 3P4
- Corresponding author: Department of Biological Sciences, University of Windsor, 401 Sunset Avenue, Windsor, ON, Canada N9B 3P4. Tel: +1 519 253 3000 ×2701.
| | - Craig E. Franklin
- School of Biological Sciences, The University of Queensland, Brisbane, QLD4072, Australia
| | - Kevin R. Hultine
- Department of Research, Conservation and Collections, Desert Botanical Garden, 1201 North Galvin Parkway, Phoenix, AZ85008, USA
| | - Mark van Kleunen
- Ecology, Department of Biology, University of Konstanz, Universitätsstrasse 10, D 78457 Konstanz, Germany
| | - Robert J. Lennox
- Fish Ecology and Conservation Physiology Laboratory, Department of Biology and Institute of Environmental Science, Carleton University, 1125 Colonel By Drive, Ottawa, ON, CanadaK1S 5B6
| | - Oliver P. Love
- Department of Biological Sciences, University of Windsor, 401 Sunset Avenue, ON, CanadaN9B 3P4
| | - Jodie L. Rummer
- ARC Centre for Excellence for Coral Reef Studies, James Cook University, Townsville, QLD4811, Australia
| | - Steven J. Cooke
- Fish Ecology and Conservation Physiology Laboratory, Department of Biology and Institute of Environmental Science, Carleton University, 1125 Colonel By Drive, Ottawa, ON, CanadaK1S 5B6
| |
Collapse
|