A hybrid Grey-Markov/ LUR model for PM10 concentration prediction under future urban scenarios
Introduction
Air pollution, especially particulate matter, has been associated with numerous adverse effects on human health in urban areas, including increased mortality and morbidity from respiratory, lung, and cardiopulmonary cancer (Zou et al., 2015; Loomis et al., 2013; Beelen et al., 2014; Wang et al., 2014; Kim et al., 2015). Though concentrations of urban PM10 generally have been declining since the turn of the 21st century (Cheng et al., 2013; Querol et al., 2014), the control of particulate matter remains an enormous challenge with increasing industrial production, travel behaviour and construction activities that follow decades of rapid urbanization (Feng et al., 2017). Understanding the spatial distribution of PM10 concentrations under varying future urban planning scenarios is a crucial challenge in designing urban development strategies and the prevention of air pollution exposure.
Various approaches to predict the concentrations of urban air pollutants have been tested. Efforts include temporal forecasting and spatial mapping. The temporal forecast makes predictions on historical relationships and trends from data on air pollutant observation. The commonly used methods are well-tested and have demonstrated promising forecasting accuracy. This is typified by neural networks, support vector machine learning, support vector regression, parametric and nonparametric regression, autoregressive moving (integrated) average modelling, grey system theory, etc. (Hooyberghs et al., 2005; Hrust et al., 2009; Kumar and Jain, 2010a; Lotfalipour et al., 2013; Qin et al., 2014; Wang et al., 2015; Donnelly et al., 2015; Hamzacebi and Karakurt, 2015). Among these, the grey method has been proved to be an effective way to make predictions of air pollutants at a relatively large scale (e.g. annual scale) under conditions with limited data. Yet forecasting precision of grey method might be affected by the random fluctuations of the data sequence. The Grey-Markov model, which introduced the Markov chain models to reduce the random fluctuation, has been successfully applied in forecasting the electric-power demand, energy consumption and foreign tourist arrivals (Huang et al., 2007; Kumar and Jain, 2010b; Sun et al., 2016). It could be an effective way to improve the accuracy of the grey method for fluctuated air pollution data sequences. However, an important mutual limitation is that all those methods are usually conducted at the station level, which is not able to characterize the patterns of air pollution under future urban scenario spatially and to evaluate the validity of varied urban planning strategies.
Spatial mapping, on the other hand, attempts to retrieve the spatial distribution characteristics of air pollution concentrations based on physiochemical processes or the relationship between air pollutants and their potential predictors. Satellite remote sensing (RS), air dispersion modelling, spatial interpolation, and land use regression (LUR) models are practical, frequently used solutions (Fang et al., 2016; Zou et al., 2016a, 2016b, 2017; Apte et al., 2017; Zhai et al., 2018). LUR, which predicts concentration of air pollutant at a given site based on surrounding land use, meteorology factors and other variables obtained through geographic information system (GIS), is now a popular method for air pollution estimation at fine spatial resolutions because of its low requirement of intensive computations and easy availability of related data (Henderson et al., 2007; Ross et al., 2007; Hoek et al., 2010; Zou et al., 2015; Jedynska et al., 2017). Due to the follow-up long-term epidemiological studies usually have longer periods than monitoring data used in LUR modelling; few investigations have tried to transfer the LUR model across time (Mölter et al., 2010; Marcon et al., 2015; Meng et al., 2015; Zou et al., 2016c). Although the results indicate that the LUR models were reasonably stable over time and it was possible to transfer them to different years, these reported studies were performed retrospectively. The fundamental reason why LUR is seldom used in future urban scenarios may be the lack of essential inputs of both air pollution observations and land use distributions.
Fortunately, the simulation of urban land-use change dynamics is well developed and comprehensive. The methods can be put into the following groups: cellular automata model, agent-based model, empirical statistical model, and hybrid models (Verburg et al., 2004; Chen et al., 2008; Santé et al., 2010; Zhang et al., 2013; Basse et al., 2014; Groeneveld et al., 2017; Liu et al., 2017). Among them, the conversion of land use and its effects at the small regional extent (CLUE-S) model can derive empirically quantitative relations between land use change and driving factors from cross-sectional analysis at multiple scales. This simulates possible changes under land use scenarios spatially explicit in small regions at a fine spatial resolution. It has been introduced with notable accuracy globally, which provides a reliable foundation for analysis of future urban planning strategies (Verburg et al., 2002).
In this study we developed a hybrid Grey-Markov/LUR (GMLUR) model to explore the spatial patterns of PM10 concentrations under future urban scenarios. Research integrates the prediction of station-based Grey-Markov modelling with the spatial mapping of LUR. The overall objective is to extend two-dimensional spatial mapping of LUR into the three-dimensional spatial prediction of GMLUR to achieve the area-based forecast of PM10 concentrations and to illustrate the potential effect of urban planning scenarios on the temporal evaluation of the spatial distribution of PM10 concentrations.
Section snippets
Framework of GMLUR and supporting methods
The hybrid Grey-Markov/LUR modelling (GMLUR) is a calibrated LUR method which forecasts PM10 concentrations based on future land use and PM10 concentrations surrogates. The process can be divided into four steps. Step 1 develops and validates the base LUR model (LURH (t)) based on the historical observations of PM10 and various geographic elements. In Step 2 we obtain the predictions of PM10 concentration for the validation year (LURH predictions (t + n)) by apply the base LUR model to
Case study
We predicted the spatial patterns of urban PM10 concentrations for four urban scenarios in the agglomeration of Changsha-Zhuzhou-Xiangtan (CZT) for the year 2020 by applying the GMLUR model to simulated land use maps for four representative urban scenarios. The base LUR model was established at the year 2013, which was then transferred to 2015 and calibrated by Grey-Markov model-based predictions to evaluate the accuracy of GMLUR.
Base LUR model fitting and validation
The proportions of built-up area, bare area, road length, and relative humidity are highly significant predictors (p < 0.05) of PM10 concentrations in the CZT in base year (2013). The finalized LURH (2013) is built as . , , and are area proportions of built-up area, bare area, and road length within the 900 m, 300 m, and 300 m buffer, respectively. The relatively
Discussion
Due to the increasing need to understand air pollution spatial patterns under different future urban scenarios, a hybrid Grey-Markov/LUR (GMLUR) model has been innovatively developed in this study that is capable of predicting the spatial distribution of PM10 concentration. As expected, the GMLUR modelling clearly shows better performance in predicting PM10 concentrations compared with the conventional LUR modelling. The RMSE (i.e. 5.50 μg/m3) and ARPE (5.13%) of GMLUR are significantly lower
Conclusions
Understanding the spatial patterns of urban air quality under future scenarios is an urgent task as rapid global urbanization continues. To predict the spatial patterns of PM10 concentration under future urban scenarios, a hybrid Grey-Markov/LUR (GMLUR) model that combines the merits of station-based air pollution forecasting and LUR based spatial mapping was proposed in this study. The GMLUR modelling demonstrates better performance in predicting PM10 concentrations compared with conventional
Acknowledgements
This work was supported by the National Key Research and Development Program [grant number 2016YFC0206205]; the National Natural Science Foundation of China [grant number 41201384]; and the Open Fund of University Innovation Platform, Hunan, China [grant number 15K132].
References (47)
- et al.
Land use changes modelling using advanced methods: cellular automata and artificial neural networks. The spatial and explicit representation of land cover dynamics at the cross-border region scale
Appl. Geogr.
(2014) - et al.
Effects of long-term exposure to air pollution on natural-cause mortality: an analysis of 22 European cohorts within the multicentre ESCAPE project
Lancet
(2014) - et al.
Simulating the optimal land-use pattern in the farming-pastoral transitional zone of Northern China
Comput. Environ. Urban
(2008) - et al.
Characteristics and health impacts of particulate matter pollution in China (2001–2011)
Atmos. Environ.
(2013) - et al.
Real time air quality forecasting using integrated parametric and non-parametric regression techniques
Atmos. Environ.
(2015) - et al.
Aerosol particle and trace gas emissions from earthworks, road construction, and asphalt paving in Germany: emission factors and influence on local air quality
Atmos. Environ.
(2015) - et al.
Satellite-based ground PM2.5 estimation using timely structure adaptive modeling
Rem. Sens. Environ.
(2016) - et al.
Theoretical foundations of human decision-making in agent-based land use models–A review
Environ. Model. Software
(2017) - et al.
A neural network forecast for daily average PM10 concentrations in Belgium
Atmos. Environ.
(2005) - et al.
Neural network forecasting of air pollutants hourly concentrations using optimised temporal averages of meteorological variables and pollutant concentrations
Atmos. Environ.
(2009)
Predictive analysis on electric-power supply and demand in China
Renew. Energy
Spatial variations and development of land use regression models of oxidative potential in ten European study areas
Atmos. Environ.
A review on the human health impact of airborne particulate matter
Environ. Int.
Time series models (Grey-Markov, Grey Model with rolling mechanism and singular spectrum analysis) to forecast energy consumption in India
Inside Energy
High spatial resolution night-time light images for demographic and socio-economic studies
Rem. Sens. Environ.
A future land use simulation model (FLUS) for simulating multiple land use scenarios by coupling human and natural effects
Landsc. Urban Plann.
The carcinogenicity of outdoor air pollution
Lancet Oncol.
Development and transferability of a nitrogen dioxide land use regression model within the Veneto region of Italy
Atmos. Environ.
A land use regression model for estimating the NO2 concentration in shanghai, China
Environ. Res.
Modelling air pollution for epidemiologic research–Part II: predicting temporal variation through land use regression
Sci. Total Environ.
Land-cover change model validation by an ROC method for the Ipswich watershed, Massachusetts, USA
Agric. Ecosyst. Environ.
Analysis and forecasting of the particulate matter (PM) concentration levels over four major cities of China using hybrid models
Atmos. Environ.
2001–2012 trends on air quality in Spain
Sci. Total Environ.
Cited by (44)
Improving air quality through urban form optimization: A review study
2023, Building and EnvironmentAn interval feature discrete grey-Markov model based on data distributions and applications
2023, Applied Mathematical ModellingPredicting resilience in retailing using grey theory and moving probability based Markov models
2021, Journal of Retailing and Consumer ServicesForecasting the concentration of sulfur dioxide in Beijing using a novel grey interval model with oscillation sequence
2021, Journal of Cleaner ProductionAn innovative coupled model in view of wavelet transform for predicting short-term PM10 concentration
2021, Journal of Environmental ManagementCitation Excerpt :With the development of computer technology, some predictive models based on machine learning are constantly produced and developed for PM 10. For instance, Kalman filter (KF) (Zolghadri and Cazaurang 2006; Choi et al., 2018), support vector machine (SVM) (Li and Tao 2018), Grey model (GM) (Xu et al., 2018), modified extreme learning machine (MELM) (Wang et al., 2017), recursive neural network (RNN) (Biancofiore et al., 2017), echo state network (ESN) (Zhang et al., 2018a, 2018b), neural networks (NNs) (de Gennaro et al., 2013; Fernando et al., 2012; Hooyberghs et al., 2005), artificial neural network (ANN) (Grivas and Chaloulakou 2006; Franceschi et al., 2018; Yadav and Nath 2019; Schornobay-Lui et al., 2019; Moustris et al., 2013; Nejadkoorki and Baroutian 2012; Wu et al., 2011; McKendry 2002; Voukantsis et al., 2011) are adopted to forecasting the PM 10. In addition to these single prediction models, there are some coupled models used to predict PM10, such as the combination of ANN, SVM, and Taylor expansion forecasting model (TEFM) (Wang et al., 2015), the combination of random forests (RFs), genetic algorithm (GA) and back propagation neural networks (BPNN) (Dotse et al., 2018), the combination of RNN and RFs (Feng et al., 2019), and the combination of ANN and GA (Antanasijević et al., 2013).
Predictions and mitigation strategies of PM<inf>2.5</inf> concentration in the Yangtze River Delta of China based on a novel nonlinear seasonal grey model
2021, Environmental PollutionCitation Excerpt :They applied it to forecast indicators of fog and haze in Nanjing. The hybrid models are also introduced for particular forecasting matters, such as a Grey-Markov/LUR model (Xu et al., 2018) and RM-GM-FFNN model (Fu et al., 2015). Some scholars adopted effective and promising techniques to modify grey models to describe the dynamic change, seasonal fluctuation, and nonlinearity of the observed series.