Comparing spatio‐temporal models for particulate matter in Piemonte

Abstract

In the last two decades, increasing attention has been given to air pollution around the world, mainly because of its impact on human health and on the environment. In the Po valley (northern Italy), one of the most troublesome pollutant is PM 10 (particulate matter with an aerodynamic diameter of less than 10 m). In order to assess PM 10 concentration over an entire region, environmental agencies need models to predict PM 10 at unmonitored sites. To choose among possible predictive models and then meet the agencies' request, we focus on the class of Bayesian hierarchical models as they provide a flexible framework for incorporating relevant covariates as well as spatio-temporal interactions. We consider six alternative models for PM 10 concentration in Piemonte region (north-western Po Valley), during the winter season October 2005-March 2006. Our aim is to choose a model that is satisfactory in terms of goodness of fit, interpretability, parsimony, prediction capability and computational costs. In order to support this choice, we propose a comparison approach based on a set of criteria summarized in a table that can be easily communicated to non-statisticians. The comparison findings allow to provide Piemonte environmental agencies with an effective statistical model for building reliable PM 10 concentration maps, equipped with the corresponding uncertainty measure.

Spatial and spatio-temporal models with R-INLA

“…meteorological and geographical variables) as well as time and space dependence (e.g. Cocchi et al, 2007;Cameletti et al, 2011a;Sahu, 2012;Fassò and Finazzi, 2011). The main drawback of this formulation is related to the computational costs required for model parameter estimation and spatial prediction when MCMC methods are used, especially in case of massive spatio-temporal datasets.…”

Two-stage Bayesian model to evaluate the effect of air pollution on chronic respiratory diseases using drug prescriptions

“…Berrocal et al, 2010b;Zidek et al, 2011). The spatio-temporal model we specify here is widely adopted in the air quality literature thanks to its flexibility in modeling relevant covariates as well as correlation in space and time (Fassò and Finazzi, 2011;Cocchi et al, 2007;Cameletti et al, 2011;Sahu, 2011). Moreover, it has been already implemented in R-INLA and validated by Cameletti et al (2013).…”

Assessment and visualization of threshold exceedance probabilities in complex space–time settings: A case study of air quality in Northern Italy

“…The time-varying covariates are obtained from a nested system of deterministic computer-based models implemented by the environmental agency ARPA Piemonte. For a complete description and preliminary analysis of the data we refer to Cameletti et al (2011). …”

“…In this paper, we focus on PM 10 , specifically modeling the probability of exceeding certain potentially harmful thresholds. Many recent PM 10 studies address the behavior of space-time trends for this pollutant (see for instance Cameletti et al (2011), Fassò andFinazzi (2011), De Iaco et al (2012) and the references therein); however the threshold exceedance probabilities and the corresponding uncertainties are less understood. Statistically, it is possible to model and estimate the probability that a specified value of a given pollutant will be exceeded and thus identify areas where the risk of exceeding such limit values is high.…”