Statistical Modeling Approaches for PM10 Prediction in Urban Areas; A Review of 21st-Century Studies

Abstract

PM 10 prediction has attracted special legislative and scientific attention due to its harmful effects on human health. Statistical techniques have the potential for high-accuracy PM 10 prediction and accordingly, previous studies on statistical methods for temporal, spatial and spatio-temporal prediction of PM 10 are reviewed and discussed in this paper. A review of previous studies demonstrates that Support Vector Machines, Artificial Neural Networks and hybrid techniques show promise for suitable temporal PM 10 prediction. A review of the spatial predictions of PM 10 shows that the LUR (Land Use Regression) approach has been successfully utilized for spatial prediction of PM 10 in urban areas. Of the six introduced approaches for spatio-temporal prediction of PM 10 , only one approach is suitable for high-resolved prediction (Spatial resolution < 100 m; Temporal resolution ď 24 h). In this approach, based upon the LUR modeling method, short-term dynamic input variables are employed as explanatory variables alongside typical non-dynamic input variables in a non-linear modeling procedure.

Assessment of the Indoor Air Quality of Akure, South – West, Nigeria

“…It can be deduced that the emitted substances were temperature, wind speed, and humidity dependent. Wind speed is a principal factor in the control of air pollution levels Grivas and Chaloulakou, 2006), wind direction plays a major role in the transport, dilution, and re-suspension of PM 10 (Harrison et al, 1997;Shahraiyni and Sodoudi, 2016), and temperature is considered as one of the strongest predictors of PM 10 concentration (Papanastasiou et al, 2007). Figure 3 showed the results of the assessments determined in this study.…”

A Practical Approach to the Real Time Prediction of PM10for the Management of Indoor Air Quality in Subway Stations

“…MLR has been used for predicting PM10 concentrations in ambient air [33]. Recently, MLR has also been used for PM10 in subways [24,30,31].…”

Twitter-Based Sensing of City-Level Air Quality

“…In the past, several researchers have tried to address the problems of estimating current air quality in unmonitored locations (spatial prediction) and short-term air quality forecasting (temporal prediction) using statistical approaches that model the relationships between air pollutants and various explanatory variables such as lagged pollutant observations, wind speed, solar radiation, cloud coverage, air temperature, traffic, etc. (see [5] for a detailed review of such methods).…”

“…Inverse Distance Weighting (IDW) and variations of Kriging [15]), dispersion models [16] and Land Use Regression (LUR) variants [17]. Among these methods, dispersion and LUR are known to generate robust long-term intra-city predictions (when enough data are available) but spatial interpolation is usually preferred for spatially coarser short-term estimations [5]. Therefore, in our study we compare our city-level Twitter-based estimates with those generated by a spatial interpolation method (IDW).…”