Environmetrics最新文献

Due to the high costs of monitoring environmental processes, measurements are commonly taken at different temporal scales. When observations are available at different temporal scales across different spatial locations, we name it temporal misalignment. Rather than aggregating the data and modeling it at the coarser scale, we propose a model that accounts simultaneously for the fine and coarser temporal scales. More specifically, we propose a spatiotemporal model that accounts for the temporal misalignment when one of the scales is the sum or average of the other by using the properties of the multivariate normal distribution. Inference is performed under the Bayesian framework, and uncertainty about unknown quantities is naturally accounted for. The proposed model is fitted to data simulated from different spatio-temporal structures to check if the proposed inference procedure recovers the true values of the parameters used to generate the data. The motivating example consists of measurements of total pollen concentration across Toronto, Canada. The data were recorded daily for some sites and weekly for others. The proposed model estimates the daily measurements at sites where only weekly data was recorded and shows how the temporal aggregation of the measurements affects the associations with different covariates.

We propose a statistical emulator for a climate-economy deterministic integrated assessment model ensemble, based on a functional regression framework. Inference on the unknown parameters is carried out through a mixed effects hierarchical model using a fully Bayesian framework with a prior distribution on the vector of all parameters. We also suggest an autoregressive parameterization of the covariance matrix of the error, with matching marginal prior. In this way, we allow for a functional framework for the discretized output of the simulators that allows their time continuous evaluation.

Air pollution continues to be a major environmental concern in China. The wind-driven transmission poses difficulties in understanding the air pollution patterns at the local level. The main objective of this study is to offer a straightforward approach for investigating the temporal trends and meteorological effects on the air pollutant concentrations during the generation process without being confounded by the complex wind-driven transmission effect. We focus on the hourly data of the three most common air pollutants: PM2.5, NO$��{}_{�2�}�$, and CO under air stagnation in Beijing, China, during 2014–2017. We find that the local pollution levels under air stagnation in Beijing have decreased over the years; winter is the severest month of the year; Sunday is the clearest day of the week. Our model also interpolates the air pollutant concentrations at sites without monitoring stations and provides a map of air pollution concentrations under air stagnation. The results could be used to identify locations where air pollutants easily accumulate.

Geostationary weather satellites collect high-resolution data comprising a series of images. The Derived Motion Winds (DMW) Algorithm is commonly used to process these data and estimate atmospheric winds by tracking features in the images. However, the wind estimates from the DMW Algorithm are often missing and do not come with uncertainty measures. Also, the DMW Algorithm estimates can only be half-integers, since the algorithm requires the original and shifted data to be at the same locations, in order to calculate the displacement vector between them. This motivates us to statistically model wind motions as a spatial process drifting in time. Using a covariance function that depends on spatial and temporal lags and a drift parameter to capture the wind speed and wind direction, we estimate the parameters by local maximum likelihood. Our method allows us to compute standard errors of the local estimates, enabling spatial smoothing of the estimates using a Gaussian kernel weighted by the inverses of the estimated variances. We conduct extensive simulation studies to determine the situations where our method performs well. The proposed method is applied to the GOES-15 brightness temperature data over Colorado and reduces prediction error of brightness temperature compared to the DMW Algorithm.

This paper draws its motivation from applications in geophysics, agricultural, and environmental sciences where empirical evidence of slow decay of correlations have been found for data observed on a regular lattice. Spatial ARFIMA models represent a widely used class of spatial models for analyzing such data. Here, we consider their generalization to conditional autoregressive fractional integrated moving average (CARFIMA) models, a larger class of long memory models which allows a wider range of correlation behavior. For this class we provide detailed descriptions of important representative models, make the necessary comparison with some other existing models, and discuss some important inferential and computational issues on estimation, simulation and long memory process approximation. Results from model fit comparison and predictive performance of CARFIMA models are also discussed through a statistical analysis of satellite land surface temperature data.

Regional climate models (RCM) describe the mesoscale global atmospheric and oceanic dynamics and serve as dynamical downscaling models. In other words, RCMs use atmospheric and oceanic climate output from general circulation models (GCM) to develop a higher resolution climate output. They are computationally demanding and, depending on the application, require several orders of magnitude of compute time more than statistical climate downscaling. In this article, we describe how to use a spatio-temporal statistical model with varying coefficients (VC), as a downscaling emulator for a RCM using VC. In order to estimate the proposed model, two options are compared: INLA, and varycoef. We set up a simulation to compare the performance of both methods for building a statistical downscaling emulator for RCM, and then show that the emulator works properly for NARCCAP data. The results show that the model is able to estimate non-stationary marginal effects, which means that the downscaling output can vary over space. Furthermore, the model has flexibility to estimate the mean of any variable in space and time, and has good prediction results. INLA was the fastest method for all the cases, and the approximation with best accuracy to estimate the different parameters from the model and the posterior distribution of the response variable.

We propose a new autoregressive moving average (ARMA) process with generalized gamma (G$��\Gamma �$) marginal law, called G$��\Gamma �$-ARMA. We derive some of its mathematical properties: moment-based closed-form expressions, score function, and Fisher information matrix. We provide a procedure for obtaining maximum likelihood estimates for the G$��\Gamma �$-ARMA parameters. Its performance is quantified and discussed using Monte Carlo experiments, considering (among others) various link functions. Finally, our proposal is applied to solve remote sensing problems using synthetic aperture radar (SAR) imagery. In particular, the G$��\Gamma �$-ARMA process is applied to real data from images taken in the Munich and San Francisco regions. The results show that G$��\Gamma �$-ARMA describes the neighborhoods of SAR features better than the gamma-ARMA process (a reference for asymmetric positive data). For pixel ray modeling, our proposal outperforms $��{�𝒢�}_{�I�}^{�0�}�$ and gamma-ARMA.

We introduce a flexible and scalable class of Bayesian geostatistical models for discrete data, based on nearest-neighbor mixture processes (NNMP), referred to as discrete NNMP. To define the joint probability mass function (pmf) over a set of spatial locations, we build from local mixtures of conditional pmfs using a directed graphical model, with a directed acyclic graph that summarizes the nearest neighbor structure. The approach supports direct, flexible modeling for multivariate dependence through specification of general bivariate discrete distributions that define the conditional pmfs. In particular, we develop a modeling and inferential framework for copula-based NNMPs that can attain flexible dependence structures, motivating the use of bivariate copula families for spatial processes. Moreover, the framework allows for construction of models given a pre-specified family of marginal distributions that can vary in space, facilitating covariate inclusion. Compared to the traditional class of spatial generalized linear mixed models, where spatial dependence is introduced through a transformation of response means, our process-based modeling approach provides both computational and inferential advantages. We illustrate the methodology with synthetic data examples and an analysis of North American Breeding Bird Survey data.

Coral reefs have become one of the most vulnerable ecosystems worldwide due to rising environmental and anthropogenic pressures. Methods from experimental design can be used to furnish our ability to monitor such ecosystems efficiently. Recently, adaptive design approaches have been proposed for monitoring coral reefs; however, questions have surfaced around the ability of such approaches to capture trends over time. The aim of this study was to develop an approach to assess trends in hard coral cover and evaluate the effectiveness of adaptive designs for estimating such trends in coral reef communities within a region of the Great Barrier Reef. Our approach was couched within a Bayesian design and inference framework such that uncertainty was captured rigorously and so that information from accumulating data can be incorporated straightforwardly to inform future data collection. The designs found under this approach were compared to historical non-adaptive designs which surveyed all locations over time. Through this comparison, we show that adaptive designs can maintain trends over time with little to no loss in information, even when sampling effort is substantially reduced. Accordingly, this research serves to further promote adaptive design methods for efficiently and effectively sampling in ecological monitoring.

Spatial maps of extreme precipitation are crucial in flood prevention. With the aim of producing maps of precipitation return levels, we propose a novel approach to model a collection of spatially distributed time series where the asymptotic assumption, typical of the traditional extreme value theory, is relaxed. We introduce a Bayesian hierarchical model that accounts for the possible underlying variability in the distribution of event magnitudes and occurrences, which are described through latent temporal and spatial processes. Spatial dependence is characterized by geographical covariates and effects not fully described by the covariates are captured by spatial structure in the hierarchies. The performance of the approach is illustrated through simulation studies and an application to daily rainfall extremes across North Carolina (USA). The results show that we significantly reduce the estimation uncertainty with respect to state of the art techniques.