Spatial and meteorological relevance in NO2 estimations: a case study in the Bay of Algeciras (Spain)
- 140 Downloads
This study focuses on how to determine the most relevant variables in order to estimate the hourly NO2 concentrations in a monitoring network located in the Bay of Algeciras (Spain). For each station of the network, artificial neural networks and multiple linear regression have been used to compute hourly estimation models. Meteorological variables and hourly NO2 concentrations from the nearby stations have been used as inputs, and a feature selection procedure has been applied as a previous step. The different models developed have been statistically compared. The inputs used in the best estimation model for each station were the most important to estimate each hourly NO2 concentration level. These estimations can be a very useful resource to provide autonomous capacities as automatic decalibration detection or missing data imputation in monitoring networks. Finally, the similarities between stations, according to the relevance of variables, have been analysed with the aid of a hierarchical clustering algorithm.
KeywordsArtificial neural networks Monitoring networks Air pollution Feature relevance
This work is part of the coordinated research projects TIN2014-58516-C2-1-R and TIN2014-58516-C2-2-R supported by MICINN (Ministerio de Economía y Competitividad-Spain). Monitoring data have been kindly provided by the Environmental Agency of the Andalusian Government.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
- Bartra J, Mullol J, Del Cuvillo A et al (2007) Air pollution and allergens. J Investig Allergol Clin Immunol 17:3–8Google Scholar
- Bishop CM (1995) Neural networks for pattern recognition. Oxford University Press Inc, New YorkGoogle Scholar
- Dominick D, Latif MT, Juahir H et al (2012) An assessment of influence of meteorological factors on PM10 and NO2 at selected stations in Malaysia. Sustain Environ Res 22:305–315Google Scholar
- European Environment Agency (2013) Every breath we take: Improving air quality in Europe. Publications Office of the European Union, LuxembourgGoogle Scholar
- European Environment Agency (2014) Annual report 2014 and EMAS environmental statement 2014. Publications Office of the European Union, LuxembourgGoogle Scholar
- Finlayson-Pitts BJ, Pitts JN Jr (2000) Chemistry of the upper and lower atmosphere: theory, experiments, and applications. Academic Press, CambridgeGoogle Scholar
- Kukkonen J, Partanen L, Karppinen A et al (2003) Extensive evaluation of neural network models for the prediction of NO2 and PM10 concentrations, compared with a deterministic modelling system and measurements in central Helsinki. Atmos Environ 37:4539–4550. https://doi.org/10.1016/S1352-2310(03)00583-1 CrossRefGoogle Scholar
- Reyes MM (2015) Modelado de alta resolucion para el estudio de la respuesta oceanica al forzamiento del viento en el Estrecho de Gibraltar (Unpublished doctoral dissertation). University of Cádiz, SpainGoogle Scholar
- Rumelhart DE, Hinton GE, Williams RJ (1986) Learning internal representations by error propagation. In: Rumelhart DE, McClelland JL, PDP Research Group (eds) Parallel distributed processing: explorations in the microstructure of cognition, vol 1. Foundations. MIT Press, Cambridge, MA, pp 318–362Google Scholar
- Sarle WS (1995) Stopped training and other remedies for overfitting. In: Proceedings of 27th Symposium Interface Computer Science and Statistics, pp 352–360Google Scholar
- Solomatine D, See LM, Abrahart RJ (2008) Data-driven modelling: concepts, approaches and experiences. In: Abrahart RJ, See LM, Solomatine DP (eds) Practical hydroinformatics: computational intelligence and technological developments in water applications. Springer, Berlin, pp 17–30CrossRefGoogle Scholar
- Willmott CJ (1982) Some comments on the evaluation of model performance. Am Meteorol Soc 63:1309–1313. https://doi.org/10.1175/1520-0477(1982)063%3c1309:SCOTEO%3e2.0.CO;2 CrossRefGoogle Scholar