Article
Environmental Sciences
Akio Tada, Haruya Tanakamaru
Summary: Accurate estimation of riverine constituent loads is crucial for water pollution control and erosion control. However, there is currently no established method for accurately estimating loads using commonly employed water quality monitoring strategies. In this study, we propose methods for unbiased load estimation and confidence interval construction based on limited samples collected with commonly used monitoring strategies. The effectiveness of the proposed methods was verified using multiple datasets, demonstrating unbiased load estimates and appropriate confidence intervals. This research is important for evaluating and improving water quality monitoring strategies to achieve more accurate load estimation.
WATER RESOURCES RESEARCH
(2022)
Article
Computer Science, Artificial Intelligence
Fabian Guignard, Federico Amato, Mikhail Kanevski
Summary: This paper introduces novel estimations to improve understanding of ELM variability by providing identification and interpretation of different variability sources.
Article
Environmental Sciences
Akio Tada, Haruya Tanakamaru
Summary: This study proposes unbiased point estimation and interval estimation methods for river loads based on the rating curve (RC) method, using importance sampling and the bootstrap method. The effectiveness of these methods is demonstrated for river loads from both small catchments and large watersheds. The results show unbiased estimates and appropriate coverage of confidence intervals, regardless of the RC model used.
WATER RESOURCES RESEARCH
(2021)
Article
Mathematical & Computational Biology
Damon M. Bayer, Michael P. Fay, Barry I. Graubard
Summary: We have developed and studied a new method for estimating disease prevalence in complex surveys with imperfect assays. Our method combines gamma intervals and sensitivity/specificity estimates to improve coverage rates. Comparisons with established methods show that our approach performs well, particularly in low prevalence scenarios.
STATISTICS IN MEDICINE
(2023)
Article
Public, Environmental & Occupational Health
Marc Y. R. Henrion
Summary: This article discusses how to address the limitations of facility- or study-based estimates by combining multiple independent parameter estimates. The bootComb R package is introduced as a tool for calculating combined parameter estimates and deriving corresponding confidence intervals.
INTERNATIONAL JOURNAL OF EPIDEMIOLOGY
(2021)
Article
Mathematics, Interdisciplinary Applications
Shu Fai Cheung, Ivan Jacob Agaloos Pesigan
Summary: There are three common types of confidence interval (CI) in SEM: Wald-type CI, bootstrapping CI, and LBCI. LBCI has advantages in coverage probabilities, hypothesis testing, computational intensity, and invariance to transformations. However, it is not widely available in popular SEM software packages. To address this limitation, an R package called semlbci is developed for forming LBCI in models fitted by lavaan, providing researchers with more options in forming CIs for parameters in SEM.
STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL
(2023)
Article
Biochemistry & Molecular Biology
David R. Bickel
Summary: Confidence intervals of divergence times and branch lengths do not reflect uncertainty about their clades or model assumptions. Uncertainty about clades can be adjusted by multiplying confidence level with bootstrap proportion, while uncertainty about the model can be propagated by reporting the union of confidence intervals from plausible models. The proposed methods of uncertainty quantification may be used together.
MOLECULAR PHYLOGENETICS AND EVOLUTION
(2022)
Article
Engineering, Multidisciplinary
Mykhaylo Dorozhovets
Summary: This article examines the joint and marginal distributions for the location and scale parameters of the double exponential (Laplace) population for different numbers of observations. By transforming the joint distribution of estimators, the distributions are obtained, with the presence of module functions in the population distribution model being a key issue. General procedures for deriving the joint distribution of estimators for odd and even numbers of observations are introduced, along with calculations for expected values, variances, uncertainties, and confidence intervals for population parameters at various confidence levels. The results obtained were validated through Monte-Carlo simulations.
Article
Chemistry, Multidisciplinary
Shichao Wei, Shaohua Li, Siyu Yu
Summary: This paper proposes a new method for quantifying uncertainty and determining the values of lithofacies proportion. A normality test was carried out to verify the applicability of the data, and the mean and variance of samples were calculated. Based on the confidence interval theory, the uncertainty of the mean value and the range of lithofacies proportion were determined. The results provide a new reference for quantifying uncertainty and serve as a data basis for subsequent modeling and reservoir research.
APPLIED SCIENCES-BASEL
(2023)
Article
Business
Sangcheol Song
Summary: This study examines how multinational corporations (MNCs) design their investment attributes under the influence of host market uncertainty and growth rates. Using data on Korean overseas manufacturing subsidiaries, the study finds that MNCs choose either more-local-embedded small investments or less-locally-embedded large investments under high demand uncertainty and GDP growth rates. The study also finds that this choice is moderated by host market political risk and competition.
INTERNATIONAL BUSINESS REVIEW
(2022)
Article
Geochemistry & Geophysics
Wan-Lin Hu
Summary: Uncertainties in geological structural geometry can have implications for seismic potential and structural evolution. Interpretation uncertainties, in particular, can impact seismic potential assessments.
Article
Computer Science, Information Systems
Changxiao Cai, H. Vincent Poor, Yuxin Chen
Summary: We study the distribution and uncertainty of non-convex optimization for noisy tensor completion, focusing on a two-stage estimation algorithm that characterizes the distribution of this nonconvex estimator. Our findings unveil the statistical optimality of nonconvex tensor completion, achieving un-improvable l(2) accuracy when estimating unknown tensor and underlying tensor factors.
IEEE TRANSACTIONS ON INFORMATION THEORY
(2023)
Article
Statistics & Probability
Daniel Rojas-Diaz, Alexandra Catano-Lopez, Carlos M. M. Velez, Santiago Ortiz, Henry Laniado
Summary: Estimating parameters and initial conditions in non-linear models is challenging. This paper proposes a method to compute a confidence sub-contour box (CSB) for inverse problems. The method is validated with epidemiological models for different diseases and offers desirable properties such as user-defined uncertainty level and identification of influential factors.
COMPUTATIONAL STATISTICS
(2023)
Article
Ecology
Luca Santini, Ana Benitez-Lopez, Carsten F. Dormann, Mark A. J. Huijbregts
Summary: This study provides predictions of average population density, their natural variability, and statistical uncertainty for 4,925 terrestrial mammal species using a large-scale dataset. The results show that small body size, fossorial behavior, and herbivorous diets are associated with higher population densities, while large size, aerial behavior, and carnivorous diets are related to lower densities. These predictions and uncertainty estimates have various applications in macroecology and conservation biogeography, such as biomass estimation, conservation assessments, and Red List assessments.
GLOBAL ECOLOGY AND BIOGEOGRAPHY
(2022)
Article
Marine & Freshwater Biology
Kalevi Salonen, Pauliina Salmi, Jorma Keskitalo
Summary: The study reveals that aggregated distribution in microscopic counting of phytoplankton is common and may lead to over-optimistic confidence intervals when estimated based on the shortcut procedure. It is suggested to approximate 95% confidence intervals using a correction factor of 1.4 and reject the recommendation of estimating intervals from the total number of counted cells. Dynamic counting strategy with real-time confidence intervals is proposed as a significant development in quality control of phytoplankton counting.
JOURNAL OF PLANKTON RESEARCH
(2021)
Article
Ecology
Luke A. Yates, Zach Aandahl, Shane A. Richards, Barry W. Brook
Summary: Specifying, assessing, and selecting statistical models are crucial for ecological research. We provide a comprehensive and accessible review on the technical aspects of cross validation for model selection, including bias correction, estimation uncertainty, score choice, and overfitting mitigation. Our recommendations include using leave-one-out cross validation or k-fold with bias correction for minimizing bias and using calibrated selection to mitigate overfitting.
ECOLOGICAL MONOGRAPHS
(2023)
Article
Ecology
Peter S. Stewart, Philip A. Stephens, Russell A. Hill, Mark J. Whittingham, Wayne Dawson
Summary: Occupancy models are commonly used in ecology to study species occurrence, but selecting among models with different sets of covariates can lead to inaccurate parameter estimates. This study investigated the consequences of collider bias in occupancy and detection processes and explored the implications for model selection using information criteria. The results showed that information criteria can accurately select detection covariates regardless of the model's purpose, but caution is needed when selecting occupancy covariates for understanding environmental variables. These findings highlight the importance of distinguishing between parameter inference and prediction tasks in ecological modeling.
Article
Evolutionary Biology
Rebecca S. Chen, Carl D. Soulsbury, Christophe Lebigre, Gilbert Ludwig, Kees van Oers, Joseph I. Hoffman
Summary: Intensive hunting activities can have major impacts on natural populations, but even recreational hunting can subtly affect animal behavior and movement. This study focused on the black grouse population in Finland to investigate the effects of hunting on genetic diversity, inbreeding, and dispersal. The results showed that while there was little genetic structure in the population, immigration rates into hunted sites were higher, suggesting that migration may mitigate the risk of inbreeding. A spatially heterogeneous matrix of hunted and unhunted regions may be important for sustainable harvests in the future.
EVOLUTIONARY APPLICATIONS
(2023)
Article
Biodiversity Conservation
Vicky Wilkinson, Shane A. Richards, Christina Naesborg-Nielsen, Scott Carver
Summary: Interventions against infectious diseases in wildlife are necessary but problematic. However, drugs can combat various pathogens while aligning with positive ethical, epidemiological, evolutionary, and socio-economic outcomes. The authors discuss how to overcome these problems and provide a framework to objectively compare the suitability of different intervention methods.
JOURNAL OF APPLIED ECOLOGY
(2023)
Editorial Material
Biodiversity Conservation
Rafael D. Zenni, Jos Barlow, Nathalie Pettorelli, Phil Stephens, Romina Rader, Tadeu Siqueira, Rowena Gordon, Thomas Pinfield, Martin A. Nunez
JOURNAL OF APPLIED ECOLOGY
(2023)
Article
Plant Sciences
Travis G. Britton, Shane A. Richards, Mark J. Hovenden
Summary: Interactions among neighbouring plants are crucial for plant growth. Common competition indices calculate the summed size of neighbouring plants or the ratio of neighbour size relative to focal plant size. However, focal-dependent indices may lead to biased estimates of neighbourhood effects on plant growth when growth is size-dependent.
JOURNAL OF ECOLOGY
(2023)
Review
Ecology
Rachele Vada, Sonia Illanas, Pelayo Acevedo, Tim Adriaens, Marco Apollonio, Olgirda Belova, Jose Antonio Blanco-Aguiar, Sandor Csanyi, Guillaume Body, Isabel G. Fernandez-De-Mera, Ezio Ferroglio, Patrick A. Jansen, Jonathan M. Jeschke, Oliver Keuling, Santiago Palazon, Kamila Plis, Tomasz Podgorski, Fiona Rickowski, Massimo Scandura, Vasilij Shakun, Graham C. Smith, Philip A. Stephens, Koen Van Den Berge, Rauno Veeroja, Stefania Zanet, Ilgvars Zihmanis, Joaquin Vicente
Summary: The American mink is an invasive species in Europe that poses a threat to endemic biodiversity and can transmit zoonotic diseases like SARS-CoV-2. In this study, we updated the distribution map of feral American minks and analyzed their temporal trends. Our findings show that the species has continued to spread across Europe in the past 15 years, increasing its potential extent of occurrence in most countries.
Article
Biology
Richard D. Gregory, Mark A. Eaton, Ian J. Burfield, Philip V. Grice, Christine Howard, Alena Klvanova, David Noble, Eva Silarova, Anna Staneva, Philip A. Stephens, Stephen G. Willis, Ian D. Woodward, Fiona Burns
Summary: This article models temporal change in species' abundance and biomass by using extensive data describing the population sizes and trends of native breeding birds in the United Kingdom (UK) and the European Union (EU). It finds significant change in the bird assemblages of the UK and EU, with overall bird abundance decreasing and losses concentrated in abundant and smaller sized species. The study also highlights the correlation between abundance trends and species' traits, migration strategy, and niche associations linked to diet.
PHILOSOPHICAL TRANSACTIONS OF THE ROYAL SOCIETY B-BIOLOGICAL SCIENCES
(2023)
Article
Environmental Sciences
Carl Chalmers, Paul Fergus, Serge Wich, Steven. N. N. Longmore, Naomi Davies Walsh, Philip. A. A. Stephens, Chris Sutherland, Naomi Matthews, Jens Mudde, Amira Nuseibeh
Summary: Birds are crucial for monitoring biodiversity and habitat health, and they play a vital role in ecosystem management. Declining bird populations can lead to reduced ecosystem services such as seed dispersal, pollination, and pest control. While monitoring birds is important for ecologists, it is often time-consuming, expensive, and challenging to manage. This paper proposes using deep learning to overcome the challenges and automate bird monitoring in camera trap data.
Article
Ecology
Sarah E. Beatham, Philip A. Stephens, Julia Coats, John Phillips, Giovanna Massei
Summary: Effective wildlife population management requires understanding species abundance. In the United Kingdom, the increase of non-native invasive grey squirrels poses a threat to native red squirrels, tree health, and the forestry industry. A camera trap index was developed to estimate target densities of grey squirrels in woodlands to guide control measures. The index had a strong linear relationship with squirrels removed in trapping operations, and results could be obtained within 6 days.
FRONTIERS IN ECOLOGY AND EVOLUTION
(2023)
Article
Ecology
Nicolas Fuentes-Allende, Philip A. Stephens, Lynne M. MacTavish, Dougal MacTavish, Stephen G. Willis
Summary: Large herbivores in highly seasonal savanna habitats often experience significant changes in body mass due to availability of critical resources and breeding events. However, documentation of this mass variation has been rarely explored. In this study, we developed a method using mineral-baited scales and camera traps to track body mass of three bovid species across wet and dry seasons in a South African savanna ecosystem. The method proved to be effective in monitoring and revealed significant changes in body mass, which were related to factors such as vegetation productivity and rainfall frequency. This non-lethal and non-invasive method has various potential applications, including informing management decisions and assessing ecological states. It may also be valuable for monitoring long-term body condition in animals affected by environmental change in arid systems.
REMOTE SENSING IN ECOLOGY AND CONSERVATION
(2023)
Article
Ecology
Jamie E. T. McKaughan, Philip A. Stephens, Russell A. Hill
Summary: Mesocarnivores are important in wildlife management, but estimating their densities is challenging. This study used camera trapping to estimate the densities of four carnivores in commercial farmland in South Africa. The study highlights the importance of careful selection of parameters and suggests the need for more studies to increase confidence in accuracy.
ECOLOGICAL SOLUTIONS AND EVIDENCE
(2023)
Article
Behavioral Sciences
F. Blake Morton, Marieke Gartner, Ellie-Mae Norrie, Yacob Haddou, Carl D. Soulsbury, Kristy A. Adaway
Summary: Urbanization is the fastest form of landscape transformation, but the understanding of the relationship between urbanization and animal behavior is still limited. In terms of foraging, bold and innovative behaviors are believed to help urban animals access and exploit new food sources. However, this study shows that there is no evidence suggesting that urban foxes are bolder and more innovative compared to rural populations.
Article
Ecology
Sarah E. Beatham, Julia Coats, Philip A. Stephens, Giovanna Massei
Summary: The study aimed to estimate the efficacy of delivering baits via feeders to grey squirrels and determine the most important factors for bait uptake. The findings showed that baits delivered via feeders can target the majority of grey squirrels in woodland environments. This implies that oral contraceptives could be a cost-effective tool to reduce the population of grey squirrels and mitigate their negative impacts on native wildlife and tree health.
Article
Biology
Kieran J. Murphy, Gretta T. Pecl, Jason D. Everett, Ryan F. Heneghan, Shane A. Richards, Anthony J. Richardson, Jayson M. Semmens, Julia L. Blanchard
Summary: The body-size relationships between predators and prey are diverse, challenging the assumptions of traditional size-based models. By studying cephalopods, researchers found that incorporating greater biological realism improved the accuracy and predictive ability of ecosystem models.