scholarly journals Spatiotemporal prediction of wildfire extremes with Bayesian finite sample maxima

2018 ◽  
Author(s):  
Maxwell B. Joseph ◽  
Matthew W. Rossi ◽  
Nathan P. Mietkiewicz ◽  
Adam L. Mahood ◽  
Megan E. Cattau ◽  
...  
Keyword(s):  
United States ◽  
Negative Binomial ◽  
Nonlinear Effects ◽  
Predictive Distribution ◽  
Fire Frequency ◽  
Size Distributions ◽  
Finite Sample ◽  
Fire Size ◽  
Posterior Predictive Distribution ◽  
Data Set

AbstractWildfires are becoming more frequent in parts of the globe, but predicting where and when wildfires occur remains difficult. To predict wildfire extremes across the contiguous United States, we integrate a 30 year wildfire record with meteorological and housing data in spatiotemporal Bayesian statistical models with spatially varying nonlinear effects. We compared different distributions for the number and sizes of large fires to generate a posterior predictive distribution based on finite sample maxima for extreme events (the largest fires over bounded spatiotemporal domains). A zero-inflated negative binomial model for fire counts and a lognormal model for burned areas provided the best performance. This model attains 99% interval coverage for the number of fires and 93% coverage for fire sizes over a six year withheld data set. Dryness and air temperature strongly predict extreme wildfire probabilities. Housing density has a hump-shaped relationship with fire occurrence, with more fires occurring at intermediate housing densities. Statistically, these drivers affect the chance of an extreme wildfire in two ways: by altering fire size distributions, and by altering fire frequency, which influences sampling from the tails of fire size distributions. We conclude that recent extremes should not be surprising, and that the contiguous United States may be on the verge of even larger wildfire extremes.

scholarly journals Wilderness shapes contemporary fire size distributions across landscapes of the western United States

Ecosphere ◽  
2013 ◽  
Vol 4 (1) ◽  
pp. art15 ◽  
Author(s):  
Sandra L. Haire ◽  
Kevin McGarigal ◽  
Carol Miller
Keyword(s):  
United States ◽  
Western United States ◽  
Size Distributions ◽  
Fire Size

scholarly journals Practical Data Synthesis for Large Samples

2018 ◽  
Vol 7 (3) ◽  
pp. 67-97 ◽  
Author(s):  
Gillian M Raab ◽  
Beata Nowok ◽  
Chris Dibben
Keyword(s):  
Longitudinal Study ◽  
Synthetic Data ◽  
Predictive Distribution ◽  
Data Sets ◽  
Posterior Predictive Distribution ◽  
Data Set ◽  
Data Synthesis ◽  
Large Samples ◽  
The Uk ◽  
Variance Estimates

We describe results on the creation and use of synthetic data that were derived in the context of a project to make synthetic extracts available for users of the UK Longitudinal Studies. A critical review of existing methods of inference from large synthetic data sets is presented. We introduce new variance estimates for use with large samples of completely synthesised data that do not require them to be generated from the posterior predictive distribution derived from the observed data and can be used with a single synthetic data set. We make recommendations on how to synthesise data based on these results. The practical consequences of these results are illustrated with an example from the Scottish Longitudinal Study.

scholarly journals A CASE STUDY ON POINT PROCESS MODELLING IN DISEASE MAPPING

2011 ◽  
Vol 23 (3) ◽  
pp. 159 ◽  
Author(s):  
Viktor Beneš ◽  
Karel Bodlák ◽  
Jesper Møller ◽  
Rasmus Waagepetersen
Keyword(s):  
Point Process ◽  
Census Data ◽  
Predictive Distribution ◽  
Population Census ◽  
Risk Map ◽  
Posterior Predictive Distribution ◽  
Data Set ◽  
Basic Model ◽  
Background Population ◽  
Tick Borne Encephalitis

We consider a data set of locations where people in Central Bohemia have been infected by tick-borne encephalitis (TBE), and where population census data and covariates concerning vegetation and altitude are available. The aims are to estimate the risk map of the disease and to study the dependence of the risk on the covariates. Instead of using the common area level approaches we base the analysis on a Bayesian approach for a log Gaussian Cox point process with covariates. Posterior characteristics for a discretized version of the log Gaussian Cox process are computed using Markov chain Monte Carlo methods. A particular problem which is thoroughly discussed is to determine a model for the background population density. The risk map shows a clear dependency with the population intensity models and the basic model which is adopted for the population intensity determines what covariates influence the risk of TBE. Model validation is based on the posterior predictive distribution of various summary statistics.

scholarly journals Ownership Variation in Violated Regulations and National Care Standards: Evidence From Social Care Providers

2021 ◽  
pp. 089976402110014
Author(s):  
Anders M. Bach-Mortensen ◽  
Ani Movsisyan
Keyword(s):  
Social Care ◽  
Negative Binomial ◽  
Third Sector ◽  
Care Providers ◽  
Data Set ◽  
Logistic Regression Models ◽  
For Profit ◽  
Care Services ◽  
Significant Difference ◽  
Research Gap

Social care services are increasingly provisioned in quasi-markets in which for-profit, public, and third sector providers compete for contracts. Existing research has investigated the implications of this development by analyzing ownership variation in latent outcomes such as quality, but little is known about whether ownership predicts variation in more concrete outcomes, such as violation types. To address this research gap, we coded publicly available inspection reports of social care providers regulated by the Care Inspectorate in Scotland and created a novel data set enabling analysis of ownership variation in violations of (a) regulations, and (b) national care standards over an entire inspection year ( n = 4,178). Using negative binomial and logistic regression models, we find that for-profit providers are more likely to violate non-enforceable outcomes (national care standards) relative to other ownership types. We did not identify a statistically significant difference between for-profit and third sector providers with regard to enforceable outcomes (regulations).

scholarly journals Dark energy survey internal consistency tests of the joint cosmological probes analysis with posterior predictive distributions

2021 ◽  
Vol 503 (2) ◽  
pp. 2688-2705
Author(s):  
C Doux ◽  
E Baxter ◽  
P Lemos ◽  
C Chang ◽  
A Alarcon ◽  
...  
Keyword(s):  
Dark Energy ◽  
Internal Consistency ◽  
Goodness Of Fit ◽  
Small Scale ◽  
P Value ◽  
Posterior Predictive Distribution ◽  
Data Set ◽  
Dark Energy Survey ◽  
Data Vector ◽  
Energy Survey

ABSTRACT Beyond ΛCDM, physics or systematic errors may cause subsets of a cosmological data set to appear inconsistent when analysed assuming ΛCDM. We present an application of internal consistency tests to measurements from the Dark Energy Survey Year 1 (DES Y1) joint probes analysis. Our analysis relies on computing the posterior predictive distribution (PPD) for these data under the assumption of ΛCDM. We find that the DES Y1 data have an acceptable goodness of fit to ΛCDM, with a probability of finding a worse fit by random chance of p = 0.046. Using numerical PPD tests, supplemented by graphical checks, we show that most of the data vector appears completely consistent with expectations, although we observe a small tension between large- and small-scale measurements. A small part (roughly 1.5 per cent) of the data vector shows an unusually large departure from expectations; excluding this part of the data has negligible impact on cosmological constraints, but does significantly improve the p-value to 0.10. The methodology developed here will be applied to test the consistency of DES Year 3 joint probes data sets.

scholarly journals Ambient temperature and subsequent COVID-19 mortality in the OECD countries and individual United States

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Costas A. Christophi ◽  
Mercedes Sotos-Prieto ◽  
Fan-Yun Lan ◽  
Mario Delgado-Velandia ◽  
Vasilis Efthymiou ◽  
...  
Keyword(s):  
United States ◽  
Ambient Temperature ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Weather Conditions ◽  
Oecd Countries ◽  
Epidemiological Studies ◽  
Sociodemographic Factors ◽  
Sensitivity Analyses ◽  
Inverse Association

AbstractEpidemiological studies have yielded conflicting results regarding climate and incident SARS-CoV-2 infection, and seasonality of infection rates is debated. Moreover, few studies have focused on COVD-19 deaths. We studied the association of average ambient temperature with subsequent COVID-19 mortality in the OECD countries and the individual United States (US), while accounting for other important meteorological and non-meteorological co-variates. The exposure of interest was average temperature and other weather conditions, measured at 25 days prior and 25 days after the first reported COVID-19 death was collected in the OECD countries and US states. The outcome of interest was cumulative COVID-19 mortality, assessed for each region at 25, 30, 35, and 40 days after the first reported death. Analyses were performed with negative binomial regression and adjusted for other weather conditions, particulate matter, sociodemographic factors, smoking, obesity, ICU beds, and social distancing. A 1 °C increase in ambient temperature was associated with 6% lower COVID-19 mortality at 30 days following the first reported death (multivariate-adjusted mortality rate ratio: 0.94, 95% CI 0.90, 0.99, p = 0.016). The results were robust for COVID-19 mortality at 25, 35 and 40 days after the first death, as well as other sensitivity analyses. The results provide consistent evidence across various models of an inverse association between higher average temperatures and subsequent COVID-19 mortality rates after accounting for other meteorological variables and predictors of SARS-CoV-2 infection or death. This suggests potentially decreased viral transmission in warmer regions and during the summer season.

scholarly journals Altitude and risk of sudden unexpected infant death in the United States

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Richard Johnston ◽  
Xiaohan Yan ◽  
Tatiana M. Anderson ◽  
Edwin A. Mitchell
Keyword(s):  
United States ◽  
Logistic Regression ◽  
Infant Death ◽  
The United States ◽  
Sudden Infant ◽  
Data Set ◽  
Live Births ◽  
Sudden Unexpected Infant Death ◽  
Increased Risk ◽  
Unexpected Infant Death

AbstractThe effect of altitude on the risk of sudden infant death syndrome (SIDS) has been reported previously, but with conflicting findings. We aimed to examine whether the risk of sudden unexpected infant death (SUID) varies with altitude in the United States. Data from the Centers for Disease Control and Prevention (CDC)’s Cohort Linked Birth/Infant Death Data Set for births between 2005 and 2010 were examined. County of birth was used to estimate altitude. Logistic regression and Generalized Additive Model (GAM) were used, adjusting for year, mother’s race, Hispanic origin, marital status, age, education and smoking, father’s age and race, number of prenatal visits, plurality, live birth order, and infant’s sex, birthweight and gestation. There were 25,305,778 live births over the 6-year study period. The total number of deaths from SUID in this period were 23,673 (rate = 0.94/1000 live births). In the logistic regression model there was a small, but statistically significant, increased risk of SUID associated with birth at > 8000 feet compared with < 6000 feet (aOR = 1.93; 95% CI 1.00–3.71). The GAM showed a similar increased risk over 8000 feet, but this was not statistically significant. Only 9245 (0.037%) of mothers gave birth at > 8000 feet during the study period and 10 deaths (0.042%) were attributed to SUID. The number of SUID deaths at this altitude in the United States is very small (10 deaths in 6 years).

Evaluation of the CONSUME and FOFEM fuel consumption models in pine and mixed hardwood forests of the eastern United States

2014 ◽  
Vol 44 (7) ◽  
pp. 784-795 ◽  
Author(s):  
Susan J. Prichard ◽  
Eva C. Karau ◽  
Roger D. Ottmar ◽  
Maureen C. Kennedy ◽  
James B. Cronan ◽  
...  
Keyword(s):  
United States ◽  
Fuel Consumption ◽  
Prescribed Burning ◽  
Forest Type ◽  
Fire Effects ◽  
Eastern United States ◽  
Validation Data ◽  
Data Set ◽  
Predicted Values ◽  
Fine Fuels

Reliable predictions of fuel consumption are critical in the eastern United States (US), where prescribed burning is frequently applied to forests and air quality is of increasing concern. CONSUME and the First Order Fire Effects Model (FOFEM), predictive models developed to estimate fuel consumption and emissions from wildland fires, have not been systematically evaluated for application in the eastern US using the same validation data set. In this study, we compiled a fuel consumption data set from 54 operational prescribed fires (43 pine and 11 mixed hardwood sites) to assess each model’s uncertainties and application limits. Regions of indifference between measured and predicted values by fuel category and forest type represent the potential error that modelers could incur in estimating fuel consumption by category. Overall, FOFEM predictions have narrower regions of indifference than CONSUME and suggest better correspondence between measured and predicted consumption. However, both models offer reliable predictions of live fuel (shrubs and herbaceous vegetation) and 1 h fine fuels. Results suggest that CONSUME and FOFEM can be improved in their predictive capability for woody fuel, litter, and duff consumption for eastern US forests. Because of their high biomass and potential smoke management problems, refining estimates of litter and duff consumption is of particular importance.

scholarly journals Development of a semi-parametric PAR (Photosynthetically Active Radiation) partitioning model for the United States, version 1.0

2014 ◽  
Vol 7 (5) ◽  
pp. 2477-2484 ◽  
Author(s):  
J. C. Kathilankal ◽  
T. L. O'Halloran ◽  
A. Schmidt ◽  
C. V. Hanson ◽  
B. E. Law
Keyword(s):  
United States ◽  
Elevation Angle ◽  
Model Performance ◽  
Diffuse Radiation ◽  
The United States ◽  
Polynomial Model ◽  
Coefficient Of Determination ◽  
Data Set ◽  
Cubic Polynomial ◽  
Solar Elevation Angle

Abstract. A semi-parametric PAR diffuse radiation model was developed using commonly measured climatic variables from 108 site-years of data from 17 AmeriFlux sites. The model has a logistic form and improves upon previous efforts using a larger data set and physically viable climate variables as predictors, including relative humidity, clearness index, surface albedo and solar elevation angle. Model performance was evaluated by comparison with a simple cubic polynomial model developed for the PAR spectral range. The logistic model outperformed the polynomial model with an improved coefficient of determination and slope relative to measured data (logistic: R2 = 0.76; slope = 0.76; cubic: R2 = 0.73; slope = 0.72), making this the most robust PAR-partitioning model for the United States currently available.

Fire frequency is relatively more important than fire size — A reply to Russell-Smith et al

2015 ◽  
Vol 192 ◽  
pp. 478
Author(s):  
Anthony D. Griffiths ◽  
Stephen T. Garnett ◽  
Barry W. Brook
Keyword(s):  
Fire Frequency ◽  
Fire Size
Sign in / Sign up

Export Citation Format

Share Document