Oil and gas development does not reduce duck pair abundance in the Prairie Pothole Region

Conservation partners are concerned that oil and gas development in the Prairie Pothole Region may reduce the abundance of breeding duck pairs using associated wetland habitat. We conducted wetland-based surveys for breeding pairs of 5 species of dabbling ducks in the Bakken oil field during 2015–2017 across a gradient of oil and gas development intensity to test the hypothesis that the abundance of breeding duck pairs on survey wetlands would decrease as the development of oil and gas resources increased. We included covariates traditionally used to predict breeding duck pairs (i.e. wetland size and class) and a spatiotemporal index of disturbance when developing zero-inflated Poisson models relating pair abundance to environmental predictors. Similar to past analyses, pair abundance was strongly associated with wetland size. Our results were mixed and suggested that the abundance of early and late nesting species was positively and negatively related, respectively, to an index of disturbance that was largely driven by oil and gas development. Regardless of the direction of the relationship, effect sizes were small and not considered biologically significant. Our findings indicate that in our study area, strategies to conserve wetland resources for breeding duck pairs should not deviate from previous prioritization metrics within the range of oil and gas development we observed. We believe that our findings may have implications to similar landscapes within the Bakken.

Greenspace Inversely Associated with the Risk of Alzheimer’s Disease in the Mid-Atlantic United States

Alzheimer’s disease (AD) is a chronic neurodegenerative disease and the most common form of dementia in older adults. Treatment of AD symptoms is very challenging and expensive. Appropriate diet as well as mental and physical activity may delay or reduce the occurrence of AD. It is unknown whether environmental factors offer potentially protective effects against the development of AD. We explored the possible beneficial effects of greenspace (trees and herbaceous cover) on the rate of AD in the mid-Atlantic US. Data for initial AD medical claims during 2011–2013 were obtained from Medicare records for 2999 ZIP codes. The percentages of land cover classes in each ZIP code were calculated based on high-resolution land cover imagery. Associations between AD and greenspace, blue space (water), and other variables were examined using zero-inflated Poisson models. The rate of AD was negatively associated with greenspace (for a greenspace increase of 10%, risk ratio (RR) = 0.91, 95% confidence interval (CI): 0.89–0.94), and blue space (for a water area increase of 10%, RR = 0.85, 95% CI: 0.81–0.89). The inverse relationships between greenspace and the risk of AD held across season, gender, and race. The rate of AD was positively associated with the concentration of fine particulate matter (PM2.5) (RR = 1.03, 95% CI: 1.02–1.05 for an increase in PM2.5 of 1 µg/m3). Our results suggest that greenspace may have protective effects for AD, although potential mechanisms are unclear and require further investigation.

Tree Height Increment Models for National Forest Inventory Data in the Pacific Northwest, USA

The United States national inventory program measures a subset of tree heights in each plot in the Pacific Northwest. Unmeasured tree heights are predicted by adding the difference between modeled tree heights at two measurements to the height observed at the first measurement. This study compared different approaches for directly modeling 10-year height increment of red alder (RA) and ponderosa pine (PP) in Washington and Oregon using national inventory data from 2001–2015. In addition to the current approach, five models were implemented: nonlinear exponential, log-transformed linear, gamma, quasi-Poisson, and zero-inflated Poisson models using both tree-level (e.g., height, diameter at breast height, and compacted crown ratio) and plot-level (e.g., basal area, elevation, and slope) measurements as predictor variables. To account for negative height increment observations in the modeling process, a constant was added to shift all response values to greater than zero (log-transformed linear and gamma models), the negative increment was set to zero (quasi-Poisson and zero-inflated Poisson models), or a nonlinear model, which allows negative observations, was used. Random plot effects were included to account for the hierarchical data structure of the inventory data. Predictive model performance was examined through cross-validation. Among the implemented models, the gamma model performed best for both species, showing the smallest root mean square error (RSME) of 2.61 and 1.33 m for RA and PP, respectively (current method: RA—3.33 m, PP—1.40 m). Among the models that did not add the constant to the response, the quasi-Poisson model exhibited the smallest RMSE of 2.74 and 1.38 m for RA and PP, respectively. Our study showed that the prediction of tree height increment in Oregon and Washington can be improved by accounting for the negative and zero height increment values that are present in inventory data, and by including random plot effects in the models.

Bayesian Spatial Joint Model for Disease Mapping of Zero-Inflated Data with R-INLA: A Simulation Study and an Application to Male Breast Cancer in Iran

Hierarchical Bayesian log-linear models for Poisson-distributed response data, especially Besag, York and Mollié (BYM) model, are widely used for disease mapping. In some cases, due to the high proportion of zero, Bayesian zero-inflated Poisson models are applied for disease mapping. This study proposes a Bayesian spatial joint model of Bernoulli distribution and Poisson distribution to map disease count data with excessive zeros. Here, the spatial random effect is simultaneously considered into both logistic and log-linear models in a Bayesian hierarchical framework. In addition, we focus on the BYM2 model, a re-parameterization of the common BYM model, with penalized complexity priors for the latent level modeling in the joint model and zero-inflated Poisson models with different type of zeros. To avoid model fitting and convergence issues, Bayesian inferences are implemented using the integrated nested Laplace approximation (INLA) method. The models are compared according to the deviance information criterion and the logarithmic scoring. A simulation study with different proportions of zero exhibits INLA ability in running the models and also shows slight differences between the popular BYM and BYM2 models in terms of model choice criteria. In an application, we apply the fitting models on male breast cancer data in Iran at county level in 2014.

Regeneration of Quercus spp. along interactive forest boundaries in a fragmented peri-urban landscape of Mexico City

SummaryBiodiversity conservation in forest fragments surrounded by a low-quality matrix requires an understanding of how ecological conditions prevailing in the matrix enter the fragments and interact with local habitat conditions. We assessed the regeneration of oak species along edge–interior gradients in forest fragments at the periphery of Mexico City. The abundance of oak saplings was sampled along transects to the forest, while the edge effect was analysed using segmented zero-inflated Poisson models for abundance data. Three oak species were dominant in terms of their relative abundances: Quercus laeta, Quercus castanea and Quercus obtusata. Regeneration of nine oak species responded nonlinearly to the edge distance, with greater sapling abundance from the edge up to 10 m into the fragment. Canopy cover and tree height decreased from edge to fragment interior, while saplings increased in open areas within the fragments (i.e., independent of edge distance). A posterior analysis indicated that Q. obtusata reacted positively to edges. These results indicate that oak regeneration is promoted by suitable habitat conditions near the boundaries. Therefore, we suggest that forest management should focus on promoting seed production and oak establishment in forest interior habitats.

Investigating Relationships Between Drinking Venues, Drinking Companions, and Corporal Punishment of Children

Limited evidence suggests that how much a parent drinks in a particular venue, such as a bar, restaurant, or a friend’s home, is associated with use of corporal punishment. However, these relationships could differ depending on their drinking companions (e.g., spouse or friends). In this study, weighted zero-inflated Poisson models were used to examine whether the relationships between venue-specific drinking frequency, heavier drinking, and corporal punishment are moderated by drinking companions in a mixed-mode sample of parents ( n = 1,599). The relationships between drinking frequency, heavier drinking, and corporal punishment varied by drinking companions, with some combinations being protective and others conferring risk. While most alcohol screening tools focus on individual alcohol use, more nuanced assessment examining where and with whom parents are drinking could be helpful in understanding risk of physical discipline.

Testing modified zeros for Poisson regression models

Excessive zeros are common in practice and may cause overdispersion and invalidate inferences when fitting Poisson regression models. Zero-inflated Poisson regression models may be applied if there are inflated zeros; however, it is desirable to test if there are inflated zeros before such zero-inflated Poisson models are applied. Assuming a constant probability of being a structural zero in a zero-inflated Poisson regression model, the existence of the inflated zeros may be tested by testing whether the constant probability is zero. In such situations, the Wald, score, and likelihood ratio tests can be applied. Without specifying a zero-inflated Poisson model, He et al. recently developed a test by comparing the amount of observed zeros with that expected under the Poisson model. In this paper, we develop a closed form for the test and compare it with the Wald, score, and likelihood ratio tests through simulation studies. The simulation studies show that the test of He et al. is the best in controlling type I errors, while the score test generally has the least power among the tests. The tests are illustrated with two real data examples.